ML20138D469
| ML20138D469 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 10/02/1985 |
| From: | NRC |
| To: | |
| Shared Package | |
| ML20138D466 | List: |
| References | |
| NUDOCS 8510240079 | |
| Download: ML20138D469 (182) | |
Text
c-o y
1 1
?
2 3
4 i
NUCLEAR REGULATORY COMMISSION 6
PUBLIC HEARING 7
PLANTATION INN g
GRANBURY, TEXAS 9
October 2 and 3, 1985 10 11 12 13 ISSUES:
1.
Basis for establishing the homogeneous hardware population for construction of 14 adequacy review.
15 2.
Stone & Webster pipe and pipe support re-analysis program.
16..
17 18 19 20 21 22 REPORTER:
Cynthia Clay 23 go2gMN 24 DO T
25
s
/
2 1
InoEx 2
Opening statement, Cindy Early Page 3 Opening statement, Vincent Noonan 4
Presentation by Ed Siskin, Stone & Webster 6
5 09en Question and Answer session 6
7 8
9 10 Exhibit No. 1 (CASE notes and questions on 6
79-14]
11 12 October 3, 1985, 8:00 a.m.
64' I3
'~
t Presentation by John Hansel, ERC 64 I4 Handout - " Objective of ISAP VII.C" 15 Collect Information on adequacy of construction work activities using 16 statistical sampling techniques.
17 Presentation by Ron Tate II Open for questions and answers.
19 Afternoon session - Statement by Vince Noonan on decision arising from 20 caucus with NRC Staff 177 21 22 23 24 25
3 1
P3Q(({glEQS MS. EARLY:
For the record, my name is Cindy 3
Early; I'm a project manager in the Division of Licensing, 4
working on the Comanche Peak Project.
This meeting today and tomorrow between the NRC and Texas Utilities is to discuss two issues.
7 One is the basis for establishing the homo-8 geneous hardware population for the construction adequacy 9
review.
10 And the second issue which we are to discuss 11 is the Stone & Webster Pipe and Pipe Support Re-analysis 12 Program.
^
13 1
More specifically, regarding the first issue, 14 the NRC staff has some questions concerning the construction 15 adequacy review program plan which we hope to settle during 16 this meeting.
17 Number one, we need your basis and rationale 18 for establishing a homogeneous hardware population.
19 Number two, we need for you to explain the 20 basisdoryoursampling,usingthe95-5criterioninselecting 21 50 items with a rejection criteria of zero.
22 Number three, and lastly, we need for you to 23 discuss the human reliability factor since this is a factor 24 in the detection or non-detection of discrepancies; and also how the sampling process will be effected by the unaccessibility 0
4 r
p 1
of some pieces of hardware.
2 Now, for the aecond issue, for_the NRC staff 3
to complete their evaluation of the as-built verification 4
walkdown performed by Stone & Webster, we need to discuss a 5
few items.
6 The first one is the recently made pipe and 7
pipe support analysis walkdown.
8 And the second one is our continuous monitoring 9
of the Stone & Webster pipe and pipe support re-analysis 10 program.
11 But, first, we would like to have the status 12 of the Stone & Webster effort, including the walkdown.
13 Now I would like to turn this meeting over to I4 Vince Noonan who has a few comments before we begin.
15 MR. NOONAN:
Good afternoon.
16'
~
~
7 d like to make a few comments to start out 17 with this morning, basically, we are here to talk to Stone &
II Webster, I guess, today, and what they're doing about the II 79-14, so the staff can understand exactly what it is, and 20 some may have questions to try to resolve.
21 Then tomorrow I understand we'll be talking 22 to John Hansel on the work that he's-doing.
N Sometime tomorrow afternoon I will have a M
caucus with the staff and its consultants; so I'd like to make sure all staff and consultants make themselves available
s 5
1 to me sometime tomorrow afternoon.
I'm not sure Qf the exact 2
time, but it depends on how this meeting gges, 3
Prior to that caucus I'll haye CASE tQ make 4
some comments to us on the thing, before we go into the caucus.
5 The purpose of the caucus with the staff is 6
basically to come back and tell you our position on the work 7
that you're doing.
8 I'd like to say, you know, either we agree 9
with what you're doing or we don't agree, you do the following, 10 et cetera.
11 I want to take that approach, and so we will 12 do that tomorrow afternoon sometime.
13 I guess with that....
.I don't have any 14 further statements.
15 John, how do you plan to proceed this morning?
16'
~
MR. BECK:
Vince, we'd like to have Ed Siskin 17 of Stone & Webster start out by giving you an overview 18 particularly slanted toward the degree of interest you've 19 expressed, specifically in the Stone & Webster program, and 20 then have it open for questions from staff members on the 21 comments.
22 MR. NOONAN:
Okay.
Ms. Ellis with CASE gave 23 us some questions here regarding the 79-14 walkdown.
I guess 24 we don't have a machine here to make copies.
25 What I'll do is introduce it into the record,
6 1
and maybe sometime today we can get copies made and pass 2
them out.
3
[The document above-referred to was marked 4
for identification as Exhibit 1, and is attached.]
5 I think the only other thing I want to say 6
to most of the--most of the things we're going to be talking 7
about, Larry Shao will be the one that will be directing the 8
NRC work on this thing, so he'll be the one responding to 9
you.
10 MR. SHAO:
Ed, as I understand, Stone & Webster 11 made a walkdown a few weeks ago to look at a few attributes; 12 you have also made a walkdown based on 79-14.
I think the 13 staff would note; how do you match the two walkdowns to use 14 for analysis?
15 MR. SISKIN:
Basically, we've only done one 16' walkdown so far, with respect to this pipe stress and support 17 analysis.
That program, that specific walkdown which will 18 be described in our project procedure, CPPP-5, covers exactly 19 what our intent was.
20 You should have a copy of that procedure and 21 the results of our walkdown by the end of next week.
22 That's in the reyiew procedure now, and we're 23 trying to get all the comments factored in on that, 24 Basically, the purpose of the walkdown, as I 25 said, in June was to review the pipe and support arrangements
y 1
in sufficient detail to convince us that it was appropriate 2
to pursue the pipe stress and support analysis, It wa.s not 3
an intent to do anything in connection with construction 4
assess program.
This was strictly in support of our efforts 5
associated with the pipe re-analysis.
6 As such, what we did, we chose a number of 7
specific characteristics that we felt were very important to 8
make sure that we had valid data on which to base our analysis.
I We chose samples of those areas, went out and 10 did the walkdown and tried to draw a conclusion of whether 11 it wqs appropriate 'to proceed with our work.
12 MR. SHAO:
Ed, as I understand, during that 13 recent walkdown you looked at four attributes, and 79-14 14 has about ten attributes.
Can you discuss one by one why the 15 other six attributes are not reported in your analysis?
16'
~
~
MR. SISKIN:
Okay.
Basically, if you consider--
I I don't have a copy of 79-14 attributes in front of me right 18 now, but the rationale we considered were, first, what are 19 the most important characteristics in making sure you have a valid pipe stress.
Obviously, the dimensions, the physical 21 configuration was very important, so we wanted to make sure 22 that we checked to see how the existing as-built drawings 23 compared to reality as far as those dimensions were concerned.
24 Pipe support locations we considered crucial; 25 the function of the pipe support, whether it was a nnn-wny
8 1
two-way, what have you, was considered crucial.
And valve 2
and support orientation was the last area that we chose.
3 obviously, we could have gone on with a lot 4
more, but we felt that that was enough ti give us a reasonable 5
assurance that the analysis we did was based on valid data.
6 When we went through the reviews, basically 7
what we found was this:
8 With respect to the valve location, this was 9
from valve to the next bend, the next branch connection, 10 the next appropriate measuring point.
We chose 80 samples, 11 based on a 95-95 sample from mill standard 105D, strictly 12 a random sample.
Of those, 79 were found in tolerance.
13 Let me digress for cne second.
When I refer 14 to tolerance, I'm not referring to any construction tolerances.
15 I'm referring to those tolerances we consider appropriate to 16 ~
show that we have a valid stress analysis.
17 For example, if you have a valve and support II location, we have used as our standard (and we went through 19 a considerable verification of the validity of that standard 20 in 1979 and reviewed that with you) where you'd have a pipe, 21 and you're talking about the valve location on that pipe--if 22 it's a length of two feet, then within three inches really 23 showed no meaningful difference in the resulting supports, M
back and forth.
25 If the dimension involved was two to five feet,
9 1
then we consider our tolerance for analysis six inches, 2
and if it's more than five. feet, we consider plus or minus 3
a foot.
That was done with a long series of analyses where 4
~
you considered the results from a valve or a support being 5
located in one point and then as much as plus or minus twelve 0
inches away from it.
So those were the tolerances we used in our 8
analysis.
Again, it has nothing to do with the construction 9
adequacy; that's a subject for other efforts.
This was 10 strictly in order to confirm that the support analysis and 11 the piping analysis was done properly.
12 In the case of the valve location we found, 13 in the 80 we checked, 79 were within our tolerances.
On 14 one three-inch chemical and volume control line, the required 15 dimension was eight feet two.and a quarter inches.
It was 16 five inches beyond our tolerance when checked.
It was, in 17 other words, one foot five inches out of position.
18 That particular case, it does not appear that 19 it would have had a meaningful effect on the analysis anyway, 20 but it was outside the tolerance we were checking.
21 In the case of the pipe support locations, 22 we checked two hundred; of those, a hundred and ninety-nine 23 were within tolerance.
There was one 16-inch core spray 24 line where it was three feet--was supposed to be three feet 25 five and a half inches to the next support.
It was, in fact,
10 1
nine and a quarter inches, or slightly under three inches 2
beyond our tolerance.
That was the reject.
Again, we do 3
not believe that would have had a substantive effect.
4 MR. HOOKWAY:
I'm Bob Hookway from Teledyne.
5 Before we get too far away from the discussion 6
on tolerances, I've seen in a number of cases with respect 7
to 79-14 types of walkdowns, and so forth, where the tolerancet 8
used for support locations and/or valve locations was typically 9
a function of diameter of the pipe in question.
10 In your case, the tolerances have no relation-11 ship whatsoever.
12 MR. SISKIN:
They're all referring only to
(
13 large bore; but once you get over the large bore question, 14 then there are no tie-ins.
15 MR. HOOKWAY:
Okay, but you could have a two-16' inch or three-inch pipe--am I right? in your large bore 17 population.
18 MR. SISKIN:
Two and a half and larger.
19 MR. HOOKWAY:
On the low end of pipes and 20 sizes, I would have a little bit of concern where we could 21 have as much as a 12-inch variation on a valve's location 22 or support location.
23 On the high end of pipe diameters your 14 tolerances are probably very restrictive in my mind, I guess, 25 And I guess that I have one concern specificall:r,
11 1
that the smaller portion of the large diameter piping with 2
respect to these tolerances seem.to be somewhat loose.
3 And one of my questions was going to be--I 4
. hope I'm not jumping ahead here--was going to be, have you--
5 and I think you've mentioned that you have justified these 6
tolerances, back in '79 when you pursued other 79-14 plans.
7 could you elaborate a little bit more on that?
8 MR. SISKIN:
Let me introduce two other 9
people here; Ron Klause is the project manager.
And 10 Allen Chan is the assistant project manager handling the 11 technical issue.
12 Let me ask Ron to answer that question.
13 MR. KLAUSE:
Ron Klause, Stone & Webster.
14 Bob, on the smaller diameter piping, we 15 didn't differentiate for the purpose of this walkdown different 16' tolerances versus pipe size because the practice on smaller 17 piping is if you do have a valve that you normally locate a 18 support near that valve because it represents a concentrated 19 mass.
20 So we felt for the purpose of confirming 21 the information to initiate our analysis, that these toler-22 ances were represented.
13 Now, as I pointed out, in the support 24 locations of this walkdown they were all within--I'm sorry--
25 one of them was f.ive inches outside of our tolerance.
12 1
MR. SISKIN:
Little less than three inches.
1 MR. KLAUSE:
Three inches, okay, 3
So from that point of view and the way we 4
selected the attributes here that we thought we didn't have 5
to differentiate tolerances versus pipe size for purposes 6
of this walkdown.
7 MR. SHAO:
Did you look at the clearances 8
between the wall penetrations where the pipes,go through?
9 MR. KLAUSE:
Larry, there's other attributes 10 that affect pipe stress analysis.
We selected the attributes 11 that are-more sensitive to the computer analysis of the 12 problem.
_(
13 We had other features in our requalification 14 effort that we are going to confirm; such things as clearances.
15 There's currently in CPPP-8, which is a field walkdown by 38 experience, pipe stress, and support engineers to check 17 such attributes as clearances or potential interferences on II piping and supports.
That will be covered in another faction II of our program.
20 MR. SHAO:
Are you going to look at the 79-14 21 attributes to see why this can be discarded, why this had 22 to be included or....
23 MR. KLAUSE:
It was not our intent when we M -
-designed the walkdown program to verify construction adequacy 25 or 79-14; 79-14 had already been addressed by the. project.
13 1
It was our purpose to just look at the data 2
that was required to initiate the stress analysis.
3 So there are attributes in 79-14 that may 4
or may not have been addressed.
5 MR. SHAo:
But on 79-14 you have to look at 6
the clearances between floors of penetration.
You're looking 7
at pipe attachment.
And all these things affect [ inaudible]
8 is my concern.
9 MR. KLAUSE:
In our data packages that are 10 being transmitted to us we have what we call a penetration 11 detail; in those penetration details it does identify the 12 as-built clearance between the pipe and the penetration.
13 Now, again, when we have our experienced analyst 14 and support engineers walk the systems down, this will be 15 one of the things that they'll look for.
16'.'
~
MR. SHAO:
My comment is you've already 17 done that 79-14 walkdown.
Why don't we match the two and II find out, see if there's any kind of problems; what's wrong 19 with that?
20 MR. KLAUSE:
I believe the purposes of the 21 two programs have different objectives.
MR. SHAO:
The 79-14 is whether it's good 23 piping or not, which.is it for you?
What's the purpose?
24 To make sure the piping analysis is right.
25 MR. SISKIN:
That's right.
We were given r
1 14 I
data and we wanted to give ourselves reasonable assurance
+mG 2
that that data was adequate:fpr us to do our analysis, 3
MR, SHAO:
My point is have you studied the 4
79-14 walkdown?
5 MR. SISKIN:
What we have here is Texas 6
Utilities has done a 79-14 walkdown, has a 79-14 program.
7 As far as I know there are no outstanding allegations con-cerning that program.
9 We wanted to make sure that, in fact, we 10 felt comfortable with the data that we were going to use 11 that came out of that program; that was the purpose, the 12 primary purpose of this program.
13 Now, there are two other facets of this 14 effort--
15 MR. SHAO:
Okay, let me ask a question; based on your analysis of the 79-14 walkdown, did you find anything 17 that's not constructed according to design, or--
18 MR. KLAUSE:
Yes, we did.
19 That is identified in our report.
It has 20 to do with the valve orientation.
21 MR. SHAO; I'm talking about 79-14, your 22 79-14 walkdown.
23 MR. KLAUSE:
We didn't do a 79-14 walkdown.
24 MR, SHAO:
I'.m talking about somebody else 25 did.
o e
15 1
MR. KLAUSE:
Oh..
I'm sorry.
2 MR. SISKIN:
As far as we understand, those i
3 discrepancies that had been found in 79-14 had been corrected.
4 I know there are other aspects going on in the CPRT program 5
that are totally separate and distinct from--
6 MR. SHAO:
Let me ask this question:
Does 7
any 79-14 walkdown may affect your analysis?
8 MR. SISKIN:
I'm not sure I understand your 9
question, Larry.
10 MR. SHAO:
If they find something wrong in 11 79-14.
Suppose they find some clearances, suppose they have 12 two inches, only have a quarter inch, may affect your 13 analysis.
14 MR. SISKIN:
If anybody finds anything, 15 whether it's Mr. Hansel's effort or any other review on-16 sit [, we're getting copies on the appropriate documentation 17 and checking to confirm that that does not adversely affect 18 our analysis.
19 MR. SHAO:
Well, what I'm trying to say, you 20 better study the 79-14 walkdown before you do an analysis.
21 MR. SISKIN:
Wait a minute.
No, I don't 22 agree with what you're saying there, Larry.
What we've done 23 is confirm that the da.ta.is, in fact, adequate to proceed 24 with the' analysis; and that's what we're doing.
25 If. there'are any discrepancies subsequently
in 1
identified, then we will address those as they're identified.
2 MR. SHAO:
But then you haye to cQnvince the 3
staff why you can neglect the six attributes.
4 MR. KLAUSE:
Larry, I believe that the documents 5
that we (Stone & Webster) has received from TUGCO already 6
represents the 79-14 effort, and I think perhaps TUGCO can 7
respond to that.
I bm. FINNERAN:
John Finneran, TUGCO.
I If there were any adverse findings in our 10 original 79-14 program, we either changed drawings to reflect 11 that condition, or we corrected the condition in the field 12 to agree with the drawings.
13 MR. SHAO:
So the drawings are all right now.
14 You all could have told me that.
15 MR. KLAUSE:
That was what we were trying to confirm.
17 MR. SISKIN:
What we were trying to do, and 18 what we have succeeded in doing--
19 MR. SHAO:
But you say based on 79-14, that 20 nothing was wrong--
21 MR. SISKIN:
I didn't say we did a 79-14.
22 MR. SHAO:
I said based on John's statement.
23 What you say is based on 79-14, you didn't 24 find anything wrong with.the drawings?
i 25 MR. SISKIN:
I don't think that's what John said.
g 17 1
MR. SHAO:
What did he say?
-2 MR. FINNERAN:
Well, I can't quote specific 3
instances, but certainly when we did our 79-14 program, we 4
.found some errors in the drawings that we had to correct.
5 MR. SHAO:
So right now the drawings are okay?
6 MR. FINNERAN:
As far as we know, they are.
17
.MR. HOOKWAY:
I think I have the same questions 8
but maybe I'll ask it a different way.
9 I think that what you've said is that you 10 are going-to' depend on other things than your CPPP-5 walk-11 down to verify the accuracy of your input data.
And'one 12 of those other_ things, I believe, is the 79-14 walkdown in 13 other TUGCO walkdowns.
14 I think I mal.. ave heard something about-15 the pending--also on the QAQC construction adequacy walkdown.
I Now if that's the case, okay, but I think 17 that you have to, first of all, understand the tolerances 18
-used in all those walkdowns and assure yourself, et cetera, 19 that those tolerances which, in a lot of cases, are 20 different-than the SWEC tolerances are acceptable to you 21 with respect to your stress analysis.
Is that clear?
22-MR. SISKIN:
Yeah, except that in this case, 23 in almost every case that we're aware of, the tolerances 24 that they're using for their walkdowns are' tighter than the tolerances we need for stress analysis.
m 18 1
MR. BOOKWAY:.
I would agree that they're 2
tighter for those attributes that you've looked at in dash 5 3
walkdown, but I don't know what your tolerances, for 4
example, are at gaps and box frames or other places.
5 I do know some of the tolerances in the 6
TUGCO walkdown, for example.
But I don't know what yours 7
are and whether the TUGCO tolerances are acceptable to you.
8 That's kind of the question.
9 MR. SISKIN:
Anything like that, tolerance 10 would be shown on the drawing.
You're talking about supports, 11 and we're doing that on a case basis and individual analysis.
12 So we would address that on a case basis, 13 as long as they are consistent with their own drawing 14 tolerances, there's no problem.
15 ge.re talking about a systematic tolerance, I0 in'that case the tolerances for construction are going to 17 be tighter than the tolerances needed for the stress analysis.
II MR. HOOKWAY:
Well, in general, I guess I 19 agree, but I'm not sure I agree with respect to box frames, 20 for example, and the tolerances that I suspect are being 21 used for a OAQC walkdown, at least.
22 I guess I'm just asking, do you understand 23 the tolerances that are being used on all this other data 24
.that you're going to rely on?
25 MR. SISKIN:
Yes, we clearly understand that.
F 19 1.
It's clearly indicat'ed.
2' Larry, do you want me to continue on with 3
the results?
4 MR.'SHAO:
We're having one question asked; 5
Has TUGCO formally reported the 79-14 walkdown results?
0 MR. FINNERAN:
We have reported formally our 7
79-14 program to the NRC, for them to approve that program, l
8 which they have.
I The results of our 79-14 walkdowns are on 10 file.
I'm not aware of any requirement that--
11 MR. SHAO:
Okay, but--
12 MR. FINNERAN:
--we publish results of the I
r 79-14.
MR. SHAO:
Okay, but if TUGCO's finding of 14 15 79-14 would not affect the piping analysis; is tha't the II..
finding?
17
[No response.]
II MR. SHAO:
You redo a 79-14, you find the I'
clearance is not right, that's not right, they are being fixed then?
21 MR. FINNERAN:
Yes; that's correct.
MR. SHAO:
And for that reason, Stone & Webster don't have to go through all the other attributes.
MR. FINNERAN:
Well, in my way of thinking, the' Stone & Webster effort was a way to show that they had
o 20 1
confidence in the information being provided to them, where 2
the analysis and the information being provided them was the 3
results of our 79-14 program.
4 The drawings that we furnished Stone & Webster 5
for the analysis are the drawings that have been through our 6
79-14 program.
7 MR. SHAO:
After--
I MR. FINNERAN:
After we had corrected the 9
deficiencies that we identified in that program.
10 MR. SHAo:
I see.
11 MR. FINNERAN:
Those were the drawings that 12 they used for their walkdown.
13 MR. SHAo:
okay.
14 By the way, our staff did some walkdowns last 15 We had some findings, maybe we'll let you know later week.
16.
on.
I Go ahead.
MR. SISKIN:
Maybe before I go on with the I'
rest of the results, maybe it's worthwhile recapping where 20 we are.
I Basically, we had the results of TUGCo's 79-14 walkdown.
We did not, nor have we ever implied that 23 we would try to repeat something, particularly since there 24 was no indication that there was any problem with it.
We 25 did, before we started our stress analysis, wanted to
21 o
o 1
develop some confidence that, in fact, that data was satis-that was the reason we did the specific walkdown 2
factory; 3
coverage in CPPP-5.
In addition, for areas not covered in 4
that procedure--
That's a mouthful; we have to find some 5
other way to....
C triple P-5.
In addition we are going.to have another
.6 7
stress review, again, basically in support of that effort, 8
which will be covered in C triple P-8.
That document will 9
be available at the end of next week; it is our plan to train people the following week and start that stress review 10 11 walkdown on the 21st.
That will take approximately one month.
12 Basically, as a result of this review--that's 13 the reason we chose the attributes that we did, the four I4 areas that we felt were most critical to make sure that we 15 had a valid pipe stress analysis.
16 We mentioned the results of that review and 17 the valve location, pipe support location, and pipe support 38 function; we checked 200 pipe supports, and all were in 19 accordance with plan.
20 The last area was valve and support orienta-In this case, we checked 200 supports and valves; only 21 tion.
22
'190 were within our tolerance, Eight were slightly outside our tolerance; in 23 other words, tolerance was plus or minus five degrees where M
valves measured by angle, and they ranged getween five point 25
o o
22 1
two and seven point 0..
In two cases there was a substantial dis-2 3
crepancy; the angle was listed as 60, and, in fact, the 4
actual angle measurement was 35 in one case.
In the second 5
case, the angle was listed as 45, and the actual angle we 6
measured was 55.
Because of the results in that particular 7
8 area, we are reconfirming 100 percent the data on pipe 9
supports and valve orientation for anything done in our 10 analysis.
11 Because of the reason, the apparent reason, 12 for this discrepancy, we drew a conclusion that it was not 13 a generic problem, it was specific to this particular area, 14 and we did not feel that any additional checks were warranted 15 as a result of that one finding.
16'
~
We felt that doing a hundred percent recheck 17 for all valves that we used for our analysis and all supports 18 we used was satisfactory.
19 Basically, that's where we stand right now 20 on this effort.
21 Ron, what would you like to add 7 22 MR. KLAUSE:
Okay; as I said before, we--
I can expand a little bit on our walkdown by the experienced 23 24 engineers.
25 MR.. HOOKWAY:
Ron, excuse me.
23 1
Can I,go back one step to the valve and 2
support orientation population.
3 You said that you filled a subpopulation, l
4 actually, within that population, and that subpopulation 5
consisted of valves with offset cps, I guess, and also 6
supports manufactured with stress, I think, is the case.
4 7'
MR. KLAUSE:
Yes.
8 MR. HOOKWAY:
By catalog component.
9 Can you elaborate on what's left from the 10 original population, and how you can qualify that?
11 MR. KLAUSE:
Yeah.
We looked at that to 12 determine essentially which supports were involved as far
~
13 as this angularity, plus as far as the walkdown according 14 to our tolerances.
15 We examined the population that we used to 16' select the samples from and of that, there were over--I II believe there were 70 supports that did not involve struts II and snubbers; and of those 70 supports none had the alignment 19 problem.
20 So we felt very comfortable in the assumption 21 that the angularity only applied to supports manufactured 22 from catalog items, such as struts and snubbers.
23 MR. HOOW AY:
I. guess I'm not familiar with M
sampling methods and procedures.
Is that then satisfactory with respect to 95-5, and mill standard 105D?
i
_-~
24 1
Is that acceptable?
2 MR. KLAUSE:
Well, again--
3 MR. HOOKWAY:
I don't know the answer; I'm 4
just asking.
5 MR. SISKIN:
Mill standard 105D doesn't really 6
cover it; basically, that was just in putting together the 7
sampling plan itself.
It's at what a rejectable population 8
was; once the population was rejected then you have to do 9
a specific review to decide what you do as a result of this 10 problem.
The one we just described is what we're doing as 11 a result of the problem and our rationale for it.
12 MR. KLAUSE:
I'm not a 105 expert, but I I3 would think that 70 out of 200 would represent a 95-95 I4 confidence.
I MR. SHAO:
Anything over 60.
I0 MR. KLAUSE:
Sixty is sufficient.
17 The answer is yes.
MR. HOOKWAY:
What was left in that population; I'
do you know? other than the struts and snubbers and valves.
0 What was left over to make up the 707 21 MR. KLAUSE:
I really can't answer that, Bob, because we took the hanger listing for all supports, and 23 those were selected randomly.
24 MR. SISKIN:
I won't swear to it, but as I 25 remember (and I will get back and confirm it) when we innkad
25 1
at.the 70,. essentially every basis. design type of hanger 2
was included in that group, whether it was box frames or 3
various other devices.
In choosing our population we just 4
basically took notice of every support, irregardless of 5
what its design was.
6 MR. HOOKWAY:
I guess that would be box 7
frames, spring hangers?
8 MR. SISKIN:
Yes.
Those two I remember 9
being covered.
10 Go ahead.
11 MR. KLAUSE:
Okay, if I may continue now, 12 there are other attributes that affect stress analysis, 13 and we have a procedure for walkdown by experienced pipe 14 stress engineers and pipe support engineers.
15 To look at some of the other attributes that may exist; some of these would be [ inaudible) and seismic 16 '
17 local reinforcements to branch connections, clearances, II to see if there's any unidentified attachments to piping.
19 We look at things such as support regidity, 20 weldi'ng details, and integral attachments.
Keep in mind 21 that when we have a walkdown by experienced people it's impossible to list every possible thing that could be 22 23 observed.
MR. SISKIN:
We wouldn't want to list every-25 thing that could be observed.
You want him to keep his eyes
26 1
open for something that.you might not anticipate but might 2
adversely af fect the situation.
3 MR. KLAUSE:
That's right.
We would be 4
depending on them and their expertise to look for just that 5
thing that Ed says, something that may have not been noticed 6
before, and certainly, to see if there might be any condition 7
in the plant that might require modification to the pro-8 cedures or the criteria that we're using in the requalifica-9 tions of the plan.
10 And, certainly, also to familiarize themselves 11 with the conditions of the piping, and the supports that 12 exist in the Comanche Peak units.
13 So, as Ed stated, we plan for this walkdown 14 to be initiated on October 21st, and prior to that the 15 engineers will be trained in the procedure for the walkdown.
MR. HOOKWAY:
How many systems do you plan 16'
~
17 to cover in that?
18 MR. KLAUSE:
Right now it's not been fully 19 established, but we're targeting approximately 50 stress 20 problems, and we are biasing the stress problems to encompass systems that will or may experience flued transients and 21 22 systems that are safety related, high-energy type systems.
23 So right now we envision walking down approxi-24 mately 50 stress problems.
25 MR. SISKIN :
Out of approximately 300 total.
w
27 1
MR. KLAUSE:.Right.
Three hundred and one.
Some of these attributes, you 2
MR. SHAO:
3 mentioned the pipe attachment, the clearance; these are 4
in the 79-14, the attributes listed in 79-14.
Somebody else 5
has done that before.
6 MR. SISKIN:
This is not to redo anything 7
that has been done with respect to 79-14.
This is to give 8
us confidence--
9 MR. SHAO:
Additional confidence.
10 MR. SISKIN:
Additional confidence.
--that 11 everything we need to consider is being considered, and that
~
12 we feel comfortable with the data we are using for our 13 analysis.
14 MR. SHAO:
Are you through?
MR. SISKIN:.We're ready to answer your questions, Larry.
17 MR. SHAO:
Okay.
I think the staff had a 18 walkdown last week.
Maybe I'll ask them to report to you 19 on that, last week.
20 MR. TERAO:
This is David Terao.
21 The staff basically walked down 10 percent 22 of the systems that were walked down by Stone & Webster.
23 What we did basically was to look at the same four attributes 24 reviewed by Stone'& Webster, and also look at other attributes 25 related to 79-14 or other attributes which could affect the
28 1
pipe and stress analysis.
Some of the attributes that we noted in 2
the systems were related to pip ~ing run geometry and that's 3
4 a 79-14 attribute; and clearances.
5 Specifically what happened was, we notice on the as-built isometric drawings that in some cases the 6
piping terminated at an anchor point which was apparently 7
But when supposed to be a containment penetration, anchor.
8 we went to the field and checked it it appeared that the 9
termination point was actually a weld to outside containment 10 and there was still about an eight-foot section of pipe 11 12 within the penetration itself and the flued head was on I3 the inner portion of the containment.
14 So we had a question about how Stone & Webster was going to treat these isometric drawings with respect to modeling the piping systems.
17 Is there something that you are aware of with respect to the penetration length not shown on the isometric II II drawings?
0 MR. KLAUSE:
Yes; we are getting details of all penetration types that are involved in Comanche Peak.
21 of those penetrations you can determine where the proper 22 anchor point is for your piping system, whether it be a 23 flued head penetration or an integral penetration with the 24 25 containment.
m 29 I
From those penetration details you can 2
determine how the piping.is to be properly modeled in your 3
stress program.
4 So it would be a combination of the penetration 5
details and the piping isometrics that would be used to 0
build the piping model.
7 MR. TERAO:
Could you answer why the eight-8 foot section of pipe within the penetration was not shown I
on the isometric drawing itself?
10 MR. KLAUSE:
I can't answer that; no, I 11 didn't make the iso.
But I would assume that for piping 12 purposes in the containment, that would be all the detail 13' that would be required: there's penetration details that I4 covers the rest of it.
MR. TERAO:
I guess my question was more 16 aimed at whether or not the pipe in the eight-foot section 17 was, say, a cast forging for the flued head, or is it 18 typically shown on all the isometrics of this eight-foot 19
. length is not part of the iso?
20 MR. SISKIN:
David, was this one case or 21 was this a repetitive situation?
22 MR. TERAO:
Well, there's one system; perhaps 23 Bob Hookway can answer tha~t.
24 We looked at tWp systems in the penetrating 25 containment.
30 1
MR. HOOKWAY:.
We first recognized that--this 2
sort of thing on one of the injection lines as it penetrated 3
containment, and what we found was the containment wall was 4
indicated which the walkdown we assumed would be an anchor, 5
and when we got there we found that actually it wasn't an 6
anchor; it was the outside con--it was the connection to 7
the flue head assembly, which is eight feet long, in essence, 8
with the anchor actually on the other end.
9 Our concern is that a stress analyst, if he 10 weren't alert enough to go looking for the penetration 11 details, then he would call that an anchor.
12 So concern came up as to will the stress 13 analyst lose eight feet of pipe.
14 MR. KLAUSE:
Oh, I can understand your 15 question now.
I0 For the penetrations the flued head in many 17 instances, in piping, depending on the pipe size, can be 18 considered as the anchor point for the pipe, but you must II include aq/ thermal expansion that exists at that anchor so 20 you'can, for your stress problems,'you can put external 21 movements to your pipe.
22 However, as far as the penetrations are con-23 cerned in the requalification effort, we are confirming 24 where the piping should be considered as an anchor.
de have established an interface with Gibson Hill who are
31 1
responsible for. qualifying the penetration to the requalifi-2 cations of our effort, and to give them the reactions from 3
piping systems from our analysis so that they can determine 4
the effects on the penetration.
5 So that is a clearly identified boundary in 6
the piping requalification effort.
7 MR. HOOKWAY:
So you do have procedures I
which will require the analyst to include the amount of I
pipe in the flued head assembly as part of their model, I 10 guess?
MR. KLAUSE:
It's addressed in the procedures 12 on how to handle the penetrations for the requalification, 13 yes.
14 MR. TERAO:
Also in the course of our reviewing 15 this Stone & Webster walkdown packages, we also noticed in the isometric drawings that there were situations where the 17 pipe penetrated a wall, not a container penetration, but 18 it could have penetrated a floor or a walled building; and there was fire retardant material between the pipe and the 20 wall itself.
21 Usually in the cases of the large bore pipe, 22 the material is quite flexible, and we wouldn't have a 23 concern about that situation.
24 But for.some of the smaller bore pipe, I 25 would say about a four-inch line, the grout apparently--the
32 fire retardant material was very stiff; apparently it I
didn't expand as much as in the cases of the large bore 2
3 pipe penetration, and we were wondering what the effect of 4
that stiffness could be on stress analysis.
5 Spacifically, what we want to know is if
~
6 you have any of the data regarding the stiffness of the fire 7
retardant material.
8 MR. KLAUSE:
Okay, I think I can answer that, 9
David, that there's an onsite group called the Damage Study 10 group; they're responsible for specifying what type of fire 11 retardant goes into what type of penetration, okay?
12 They have established a penetration sealing 4
13 schedule that identifies the material that's in there.
14 This then is being looked at for piping 15 systems that are involved, that have penetrations involve-16' And we know the limitations of the material as far ment.
17 as its effectiveness to be a fire barrier retardant, and 18 we know the allowed deflection that it can have.
19 Our piping model and our procedures are set 20 up_to give'.the ahalyst specific instructions on how to
.21 model.these penetrations involving this material in other 22
. piping models, so that the effects of this material can 23 be reflected in the piping analysis.
I 24 MR. TERAO:
I guess the question that I would 25 raise is, I don't believe the isometric drawings showed
.__g.,,
33 1
where the material was used, 2
MR. KLAUSE:
That's right.
There's a separate 3
schedule, it's called the penetration sealant schedule.
4 MR. TERAO:
So you're saying that the piping 5
stress person, whoever is modeling the system, would also 6
have those drawings, too?
7 MR. KLAUSE:
That's right.
O MR. TERAO:
That would show where this I
marking--
10 MR. KLAUSE:
That's part of the required 11 input.
I MR. TERAO:
Similarly there was a case where--
13 I believe it was the CVCS (Chemical and Volume Control I4 Systems) small bore line that came off the CVCS header and 15
-it also penetrated the wall, concrete wall, but instead of 16 using the fire retardant material it was actually grouted 17' into the wall.
And, again, this would act as an anchor on 18 the system and we're wondering how this--
And it was not 19 shown on the isometric drawing.
And we are wondering how 20 this would be picked up by Stone & Webster.
21 MR. KLAUSE:
I would assume that it would 22 show up on the sealant schedule and if it is, indeed, grout, 23 it would be considered an anchor and the analysis would 24 reflect that condition.-
It would be modeled in as an 25 anchor.
-.r
34 1
MR. HOOKWAY:
On this sealant schedule I 2
may have missed it; it says, "How are they determining 3
what the material is?"
Is there something in the process 4
that they were checking here?
5 MR. KLAUSE:
I'm not sure what their specifi-6 cations are; I'm certain that the temperature of the pipe 7
has an effect on what type of sealant that they used.
The I
high temperature line wouldn't have the rigid stuff it 9
would have.
10 MR. HOOKWAY:
What we notice, and I'm not 11 sure this is correct, but what appeared to be the case was 12 when the angular space around the pipe was small, the 13 material was very hard.
14 And if the space was very I.arge, the fire 15 retardant material remained flexible.
But if it was small 16_
it seemed to get very hard and provided restraint to the 17 system, if you wish.
18 MR. KLAUSE:
But, again, our procedures 19 would have found that problem because in stress runs it 20 is first considered rigid.
21 MR. HOOKWAY:
22 MR. KLAUSE:
In the stress run, yes.
23 And you determine the effects of the piping 24 from that rigid restraint.
25 MR. HOOKWAY:
Is that dash six or dash seven?
e
35 1
MR. KLAUSE:
That's in seven.
2 MR. SHAO:
Shouldn't fire retardant be con-3 sidered rigid?
4 MR. KLAUSE:
For purposes of analysis, to 5
determine the effects on your piping, okay, and then it's 6
rerun without the fire retardant material there.
7 If the deflection is greater than a tenth of 8
an inch, then a boot as specified by the Damage Study group S
to protect that penetration.
10 MR. SHAO:
Would you do the same for [ inaudible]
11 analysis and [ inaudible] analysis, both?
12 MR. KLAUSE:
That's correct.
I MR. TERAO:
Also related to clearances were 14 several examples where we noticed that the pipe clamp was 15 quite close and sometimes touching an adjacent support.
Again, these are not shown on the as-built 17 drawings.
Is there another vehicle that addresses this?
18 MR. KLAUSE:
Yes; this will be covered by 19 CPPP-8.
When experienced analysts walk through the plant 20 they will be looking for items just such as that that may 21 have an impact upon the analysis.
22 Clearances between pipes and equipment is one 23 of the attributes that the'y will be looking for.
24 MR. TERAO:
Wha't I,'ve addressed so far were 25 basically what I would call as-built type of attributes.
1
36 1
In reviewing some of the supports, we also 2
looked at what I call engineering type of attributes; these 3
are the type of attributes that could affect your pipe 4
stress analysis, maybe from a functional standpoint.
And 5
I'm sure you're familiar with some of these issues, such as 0
stability and restraint stiffnesses, the treadment of 7
anchorages, such as Richmond inserts, support [ inaudible]
excitation.
These type of attributes, I believe in the 9
CPRT program plan, said that they were going to be evaluated 10 by Stone & Webster prior to initiating stress analysis.
11 MR. SISKIN:
Prior to completing the stress 12 analysis.
13 MR. TERAO:
I believe you had three categories.
14 One category was going to be before starting analysis; one 15 would be during analysis, and the third perhaps after analysis.
I don't exactly recall.
17 MR. SISKIN:
I don't remember any category 18 that said before we started the analysis.
There were some
' before we started the pipe support analysis, and some before 20 we completed the. pipe support analysis, and some before we 21 completed the whole program.
22 But we're not holding up some of the pipe 23 stress effort for anything at this point.
24 MR. TERAO:
Clearly the issue of, say, 25 restraint stiffness, has to be resolved before you start
~ - - -, -,
a 37 1
pipe stress analysis.
2 We're wondering what the status of that 3
issue was at this time?
4 MR. KLAUSE:
Again, our design criteria, 5
CPPP-7, is written as such.
The analysis can be initiated.
6 For areas that are not resolved, that will be addressed 7
later.
All work being done currently to the procedure is 8
being marked " confirmation required," and will be gonc 9
back and reviewed for anything that may have been outstanding 10 that affected that analysis or support.
11 MR. TERAO:
Okay.
12 I understand what you just said was CPPP-7 13 is going to address how, say, the issue of restraint stiff-14 ness will be handled.
15 I think what the staff has to do is not only 16' review CPPP-7, which may tell you how that issue is being 17 treated in the stress analysis, but we need to get into the 18 basis for whatever procedure on restraint stiffness is put 19 into the dash seven procedure.
20 I guess I'm trying to find what--how we can 21 go about reviewing the basis for some of these issues that's 22 been around for several years, 23 MR. KLAUSE: Well, certainly, it's our intention 24 that once CPPP-7 is issued formally that we'll certainly 25 have a document for your perusal and for your use, and we
38 o
1 will be willing to sit down and discuss anything in there 2
that you may have questions on, on how we arrived at 3
decisions that we chose.
4 MR. TERAO:
Has the dash seven and dash six 5
and seven procedures been issued yet?
6 MR. KLAUSE:
They've been issued for comment.
7 MR. SISKIN:
They have been drafted; they 8
have not been issued.
9 MR. TERAO:
Do you have a schedule on that?
10 On both those procedures?
11 MR. KLAUSE:
Currently the schedule for CPPP-7, 12 I believe, is October the lith.
13 For CPPP-6, it's around November 1st.
14 MR. TERAO:
I believe the last time we had a meeting at Stone & Webster offices at the end of August, 16-you had told us that the pipe stress analysis would be in 17 full production by October 1st.
Is that what is happening 18 now?
Are you in full production with the pipe and stress I'
analysis at this time?
20 MR. SISKIN:
We are basically in full productio n 21 using the draft proedure, all preliminary results all 22 appropriately marked, to indicate the particular document 23 being used to reform.
Obviously, once procedure No. 7, 24 procedure No. 6 are approved, those are all going to have to 25 be fully reviewed again to confirm that any changes that o
39 1-were made since the time the work was done are properly 2
factored in.
3 MR. TERAO:
Could you approximate how many 4
systems you've started on at this time?
5 MR. SISKIN:
About a hundred and five out of 6
three hundred problems.
7 MR. SHAO:
That's on Unit l?
8 MR. SISKIN:
Unit 1.
9 MR. TERAO:
Just going back to the QAQC as-10 built verification effort that's currently--or it has been 11 just performed by John Hansel's group, was there any informa-12 tion obtained from that walkddwn as needed by Stone & Webster 13 to assure the appropriate modeling of the piping system?
o_
14 MR. SISKIN:
What we are doing is we are on 15 the distribution for each finding of Mr. Hansel's group; each 16' case, each finding, is reviewed against the specific problan 17 to determine if, in fact, it has any impact or not.
II Where it's an after-the-fact thing where we 19 have already run it, and the finding is out of tolerance, 20 but would not affect in the conclusion of the stress engineer 21 have any meaningful effect on the analysis, we would not 22 rerun it.
23 Where the :information is obtained beforehand, that's the basis for the run that we would make when we
'25 factor that information in.
45 E
1 MR.. TERAO:
Is there any specific information 2
that you are aware of that migh.t have been identified in
'3 the QAQC walkdown that should be reflected in this Stone &
4 Webster analysis?
5 MR. SISKIN:
Nothing that we've seen so far 6
materially affects the Stone & Webster analysis.
~
7 I don't know that we've seen all of them yet, 8
or we've reviewed all of them yet, so it's not an absolute 9
statement.
10 MR. SHAO:
During the last meeting I think 11 the staff said maybe you should take a look at SSCR No. 11, 12 the QAOC walkdown of supports, and finding the effects of 13 your analysis.
14 Did you do an evaluation of that?
15 MR. KLAUSE:
SSCR 11 is being reviewed at this 16..
time.
I haven't heard the results of that yet.
17 MR. SHAO:
You're in the process of reviewing 18 SSCR No. 11?
19 MR. KLAUSE:
That's correct.
20 MR. CALVO:
Wait.
Stone & Webster?
You're 21 reviewing SSCR No. 11?
22 MR. KLAUSE:
Yes, that's correct.
23 MR. CALVO:
As relates to the QAQC7 24 MR. SHAO:
The'No. 11 has--
25 MR. KLAUSE:
As it relates to our recualification.
41 1
MR. SHAO:
Pipe analysis.
l 2
[Both Mr. Shao and Mr. Calvo were speaking 3
at once, therefore, both inaudible to the reporter.]
4 MR. SHAO:
QAQC did a walkdown on pipe 5
, supports and they have some findings.
6 So the staff thinks maybe they should take 7
a look at it, No. 11, how it affected--
8 MR. HOOKWAY:
To reiterate maybe one of the 9
points that Dave has made.is that as a result of our brief 10 walkdown last week, one of the concerns we came back with 11 was that of other interferences, such as structural steel, 12 other pipe supports or other pipe situations.
13 And you did mention that dash 8 walkdown 14 would address such things.
15 The problem I see with that is that dash 8 16' is not a complete walkdown, number one.
17 And I was concerned about the remaining 18 250 or so systems where we may have some interferences 19 that could be critical.
20 MR. SISKIN:
Bob, let me interrupt.
I think 21 there's a misconception here that I ought to put straight.
22 Many of the things we're doing now are to 23 identify as early as possible where there may be a problem, 24 we may need additional actions to make sure the thing is 25 done logically and properly.
k
42 1
For example, on the item you're talking 2
about, the fact that procedure No. 8 is not a hundred percent, 3
that, again, is to make people alert to potential questions, 4
particularly in view of, in other words, the more unique 5
or unusual features of this plant.
6 our procedure is, and has been on our other 7
plants as well, that when the analysis done, when you have I
a sold system, then the stress analysis engineer with iso 9
in hand then walks the system.
That's a hundred percent.
10 That's a normal part of the routine work, once the job is 11 done.
12 So if some of these things don't get picked I3 up in interference or something like that, then we have our 14 normal last-minute "we've got to go make some changes or I
some fixes."
16 But we found that because of all the inter-17 relationships that are appropriate, that final walkdown is 18 what checks for a lot of these things, and it makes it hard 19 at the end, but that's the best way to make sure that some-20
~ thing you do on another system tomorrow doesn't adversely 21 affect this one.
22 MR. SIIAO:
I'm asking the same question; the 23 79-14 walkdown is supposed to pick up all this.
So if you 24 do a good job, the other things you're talking still hasn't 25 picked up, right?
o e
43 1
MR. HOOKWAY:
It's a concern of relativity 2
somewhat, because if it were my decision, I might spend my 3
time walking down looking for interferences such as this, 4
rather than looking for valve orientation problems.
You 5
know what I'm saying?
We're looking at every valve orienta-0 tion.
I'm wondering if it could be....
7 MR. SISKIN:
Okay.
Look at it from an I
engineer's standpoint.
You're talking about dimensions.
9 You're talking about orientations.
These will affect the 10 stress in the pipe, determined by the stress run.
It will II determine the loading on the various supports.
It's a 12 sequential affair.
13 These are problems that will have a long cas-14 cading effect on--as a result of something being incorrect.
15 16..
So these are things you're particular,1y anxious to make sure you're as right as possible up front, because to 17 redo it would be a major effort.
18 If we find an interference someplace later on, 19 the only way we're going to be sure that we found it is 20 essentially when the system is done anyway, because if you 21 did a check now an interference may be added as changes are 22 made, or as other work goes on in the plant.
You'd have 23 to recheck it anyway.
Again, the fix at that point to restore it to 25 exactly what you want and ex'actly what is acceptable is more
i' e
44
__m_
1 appropriately done at the end..
So that's the reason you 2
do it at the end.
3 It doesn't make any Sense to do something 4
today that you're going to have to repeat anyway.
5 MR. HOOKWAY:
I have a similar concern on 0
gaps and box frames, for example, the same level.
~7 MR. SISKIN:
But, again, that's something 0
that would be checked eventually.
I MR. HOOKWAY:
In a general sense, we did 10 spend four days at the site, and we did find some nice
~II things, too.
I'd like to mention those, just so you don't 12~
~
get the wrong impression.
I MR. TERAO:
Bob, I think before we go into I4 that--
15
... laughter...
I0 MR. TERAO:
--we'll caucus and then we'll 17 present our final conclusion.
MR. SHAO:
Yeah; we'll talk tomorrow.
I' MR. HOOKWAY:
That's all I have.
MR. SHAO:
Okay.
21 I think the next thing is we want to know--
22 We had a meeting in New York about a month ago--what you 23 have done in the last month, what the progress.
24 MR. KLAUSE:
Okay.
In the last month we have 25 moved closer to finalizing the design criteria and the
4 O
45 1~
procedure that we'll use to haplement the criteria, to 2-the point where those will be issued on the dates that I 3
told you earlier.
4 I believe that's a significant achievement.
5 We have, for all intents and purposes, com-pleted our report for the walkdown and that was provided.
We have identified approximately four other 8
project procedures that have to be written, and they are 9
in various stages of completion now.
10 The analysis as stated earlier has started 11 on a hundred and five systems.
There is a considerable effort 12 involved in that.
We have a number of people working on 13 those problems.
14 We believe we have an understanding of the 15 technical issues that affect our analysis, and we're approach-16'
~
~
ing the finalization of how those will be resolved in the 17 requalification.
18 This has been a very active month since we 19 last talked to you.
20 MR. SISKIN:
Based on some of the information 21 we gave you then, I think we lost probably about five days 22 on our schedule.
I think it's been a productive loss of 23 five days that we've gone into some more detail on some 24 of the outstanding issues, what's necessary to put them to 25 bed.
But it's still recoverable on the end result.
It won't
46
,o 1
affect the end dates that we talked about in the past.
2 MR. SHAO:
One thing I remember is that 3
CPRT is supposed to come out with an audit; the issue may 4
affect--
I think CPRT is supposed to come with a list of 5
all the issues, including SCR No. 7, 8, 9, 10, whatever, 6
as it may affect [ inaudible].
Has CPRT done that or not?
7 MR. SISKIN:
We have not received it from 8
CPRT yet.
Where it stands, I have to defer to CPRT.
I We are not awaiting CPRT's list, obviously, 10 to resolve the issues that we're aware of.
11 MR. SHAO:
Well, one thing I think we want 12 to know, how does CPRT audit--does CPRT have an oversight 1
13 of the system?
How do you perform this function?
I4 MR. LEVIN:
From routine literature of l
15 the design adequacy program.
To my left is Tim Snyder who I0 l
is leading our support verification.
I7 I could give you a brief overview of that, l
18 Larry.
It really breaks down into four principal areas.
I' First, the--just the general area issues, the classification of these issues.
It's related to your 21 question; I'll get back to it in a minute.
l The second, the area of action planning and 23 development, the formulations, initiative, to address these l
24 issues.
l Third, procedures and their implementation.
l i
i r..
=
47 1
And fina11yf conclusions.
2 With respect to. issues, we have been in 3
the process and nearing completion of an effort to review 4
literally thousands of pages of external source documents 5
to identify--
It's our objective to identify all external 6
source issues, not only within the piping and supports area, 7
but in all areas that may have a'pplicability to the design 8
of Comanche Peak.
9 These issues are being captured and inserted 10 into a tracking system that is automated.
II We are now encapturing the issue by name, 12 trying to characterize it as to the nature, from a technical 13 point of view as expressed by the source, as well as 14 establish disposition of that in the sources--whether he 15 considered it still open or closed, and in many cases where we have some commonality similar issues have been identified 17 by several sources, and these things are being aggregated.
18 We are taking these issues and classifying 19 them for purposes of--primarily for purposes of trending 20 and cause evaluation, for publishing our results.
21 One of the things I want to point oue is 22 that we are taking a look at the issues from the standpoint 23 of their cause and generic applicability, primarily from 24 the standpoint of any lessons that we may learn from these 25 issues and how thov may affect ontonein1 aincip14nn
a 48 1
I think that through the efforts of the 2
project of Stone & Webster, and this re-analysis effort, 3
we're not really--in view of the fact that we're not redoing 4
a hundred percent sampling of the area, that we're not 5
really concerned about the generic applicability in the 6
sense of providing a new analysis of record.
7 So we're really concerned about the potential 8
applicability elsewhere.
So we'll be looking at that.
9 We plan to--I guess at this point in time, 10 they're of the order, and I'm not sure of the exact number, II 300 issues that are logged into the system.
The evaluation is not complete; we will be transmitting them formally over to Stone & Webster; however, I wanted to point out that these issues have been available and is available to the 15 But we're going to formally transnit that.
project.
16 I don't think it's been a hindrance to the 17 effort to date, primarily because they have started with 18 a list of what are most probably the most substantive issues.
They exist in several forms.
One was a IAP report and 20 all of this information really served as input into the 21 formulation of the action plan that you've seen.
22 So at least in terms of those issues and creating a point from which to depart, I think that fair 23 24 volume is on the, table.
25 Our effort--it took relatively little effort
r-49 I
to get.that far.
I would say that we spend five percent of 2
our time probably capturing 95: pe.rcent of our issues; and 3
[
we spend the remaining 95 percent of our time insuring 4
completeness and getting the remaining five percent.
That's 5
the process that I think we're in right now.
Just absolutely going through every piece of paper we can find to insure 7
l completeness of that list.
8 MR. SHAO:
Aside from SIGNA and aside from 9
similar disposition issues, there are quite a few issues 10 i
l relating to pipe and pipe supports also.
I want to make i
11 l
sure that Stone & Webster get these issues.
t l
12 MR. LEVIN:
We're going through the ASLB i
L 13 l-records, and all information from any external source for l
14 input into the system, and we will make them aware of issues 15 that possibly they may not be aware of today.
16 i
More important than that, we're going to be i
17 taking a look at the plans for evaluating those issues, 18 tracking that process.
I' 19 j
MR. SRAO:
What's your timing to,give it to 20 Stone & Webster, these external source issues?
21 MR. LEVIN:
These external source issues, 22 some of them are available right now, but they'll be l
23 formally transmitted very shortly, I expect that there'll 24 be several that will follow thereafter, you know, as this 25 l
i
F 50 i
i process continues, some will be added.
2 Like I said, right now the numbers of the 3
order are 300.
I would expect that as we complete the 4
process there will be a few more.
5 The important thing is that in the process 6
finding those that are actually the same concern have been 7
expressed by several parties.
That is not complete yet.
8 MR. NOONAN:
I'd like to ask a question on 9
that.
As you progress down this path, will there be a check-10 list or a matrix that you can point to, like a concern for 11 No. 1, "here's where it's addressed"?
12 MR. LEVIN:
Yes.
Through the system, it will 13 be source by source, and one can, you know, identify it--cut I4 it that way as well as many other ways to go.
MR. TERAO:
Howard, excuse me.
There's going 16 to be a meeting sometime at the end of this month between ICE and CPRT to talk about some of those--the CPRT effort, 18 and specifically in the piping and pipe support area.
I' If Stone & Webster is now in full production 20 of the piping stress analysis, I'm just wonderir$g whether I
waiting towards the end of this month to discuss some of these issues that could affect the stress analysis, whether we should put of f that meeting to that late, Was there 24 some reason why it was going to be--
25 MR. LEVIN:
I'm not certain which meeting L
51 f'
1 you're referring to.
l 2
MR. TERAO:
I believe this meeting--this was 3
a meeting with the I&E and Don Norkin and his group.
4' MR. LEVIN:
Okay.
There is an audit planned, 5
an inspection audit of our files planned the week of the I'
28th of October.
The primary purpose of that is to review 7
some of our work in the area of scope, validation and design I
r adequa.cy program.
I MR. TRAMMELL:
Charles Trammell.
I'd just le like to correct--offer a correction.
There may be some II audits planned but there's no meetings currently planned 12 between I&E; if there are such meetings they will be coordi-I nated through the project and announced, and so forth.
t l
14 There may be some audits to crop up, but 15 there's no meetings planned.
16 MR. LEVIN:
Now, as to time, there may be some 17 source of confusion.
It's my understanding that the audit 18 will focus, in addition to the validation, on some of the j
checklists that have been developed for purposes--for use....
~
20 At that time some of the checklists that Tim j
Snyder has been developing that he'll be using to evaluate i
22 I
some of Stone & Webster's work will be available.
I l
23 l
Are we communicating?
l 24 l-MR. SHAO:
My original comment was, I think 25 about a month ago, I think Stone & Webster was not aware of
52 1
audits in the source issues.
2 But I want them.to know as soon as possible 3
all issues so the recap will be correct.
4 I think all should be given to Stone & Webster 5
so they can do a re-analysis properly.
6 MR. LEVIN:
I think we're right around the 7
corner from the standpoint of providing a complete record.
8 In my view, they have available to them and have incorporated 9
into their program some of the major issues that are on the 10 table.
11 What we're interested in, as I said, is that 12 last five percent to be sure nothing falls through the 13 cracks.
Very shortly that will be available.
14 MR. SHAO:
Okay, can we mention a little bit 15 your oversight function also?
16'
~
MR. LEVIN:
I'll go into that right now.
17 THE REPORTER:
Please, just a minute.
I 18 must get a microphone to you.
I'm having great difficulty.
19 MS. EARLY:
Could we take a break for a 20 couple of minutes right now?
Would that be okay?
21 MR. NOONAN:
Yes.
22
[Short recess.]
23 MS. EARLY:
I'd like to say that we did not 4
get the amplifying microphones in as we had wished.
These 25 microphones do not amplify.
'l In order for her to hear and for us to hear
(
2 each other, we're goin,g to have to speak up a little bit.
3 MR. SHAO:
Shout.
4 MS. EARLY:
Whatever.
5 As we draw this to a conclusion, we understand I
that Stone & Webster is not going to be here tomorrow.
Is 7
that correct?
1 I
MR. SISKIN:
We had no intention of being.
I We can if you like.
le MR. SHAO:
I suggest maybe at least one 11 person is here, because we may have some questions with John 12 Hansel related to pipe and pipe supports.
13 So maybe I suggest one guy from Stone & Webster I4 be here tomorrow.
MR. SISKIN:
I will come.
MS. EARLY:
Okay.
17 And before I turn this back over to Larry 18 Shao for his finishing questions, I would like to also say I'
that we're going to give Mrs. Ellis a little bit of time 0
at the end of the meeting for her statement.
21 Larry.
MR. SHAO:
I think Howard Levin was in the 23 process of talking.
24 MR. LEVIN:
Okay.
I'll continue, then.
25 I'd left off with a description of how we're
-.,, - -..,., - - ~
-...--,----,..,--,e.
i 54 1
dealing with issues., and I want to proceed into our overview I
2 activities, particularly.in two areas, and the first being 3
that of action plans.
4 And I would characterize our effort in that 5
regard as participating jointly in the overall formulation 6
of approach on the one hand, and that activity led to the 7
discipline specific action plan that you fellows have seen 8
in Revision 2 of the program plan.
9 A more detailed level of execution, stone 10
& Webster and the project are formulating technical positions 11 to resolve these issues.
There has been contact and comments 12 and an in-process overview sense in that process; as well as 13 a process of review of these various CPPP procedures.
I4 That is in a time frame before these things 15 are finalized.
At the stage that they are finalized we plan 16 to subject those documents and technical approaches to more I
formalized review to be, governed by a checklist, and it I
will be that documentation that will essentially be--support II our final. conclusion with regard to the adequacy of the procedures, adequacy of the technical resolutions, and also 21 we plan to do a selected overview of the specific implementa-tion.
23 our documentation that flows out of that will be input into a results report that we will write, that 25 will provide our conclusions with respect to our investigation
~
o.
55
., ~..
~.. -. - -
1 of root cause and generic implications and the potential
~
2 applicability elsewhere to other. design.
3 Secondly, the evaluation to the quality of 4
the piping and support designs themselves.
5 And lastly, an overall assessment of the 0
completeness and technical adequacy of the Stone & Webster 7
and the project effort.
I That's a capsule of our review.
If you have 9
any other questions, I'll be glad to answer them.
10 MR. NOONAN:
I'd like to ask a question of 11 Ed Siskin.
I know about all the stuff you've been lo6 king 12 at and all you've been doing to date.
^
13 I'm looking back, maybe a little bit, on 14
-your experiences, on other--not this plant, but other plant's, 15 I'm wondering if you've found any "oh, my God" yet here, 16 and what you've seen here compared to other plants.
17 MR. SISKIN:
Actually, in all the efforts 18 that we've done to date, we have not found any "oh my Gods,"
19 not at all.
20 Based on many of the issues that have been 21 raised we really expected to find more significant things 22 than we have to date.
So far I'm very encouraged.
23 Obviously, what happens tomorrow may change 24 that, but I see no indication of a problem in that direction.
25
56 1
With respect to 79-14, we've been involved 2
in 79-14 for a considerable number of plants well in excess 3
of 15 personally.
The 79-14 effort on this plant compares 4
very favorably to any I have seen anywhere else.
That 5
comment refers specifically to 79-14, adequacy and documenta-6 tion.
That doesn't mean it is perfect; there is no plant 7
that is perfect.
But it is in very good shape for what 8
we've been able to determine so far.
9 MR. SHAO:
Maybe a followup of Vince's 10 question, all the support can be analyzed, are any support can 11 not be analyzed?
12 MR. SISKIN:
We haven't looked at all of 13 the supports yet so I can't really answer that question, 14 Larry.
15 I'd rather defer an answer to that question 16 until we've looked at least at all the supports, and until 17 obviously we've addressed the stability question, the 18 stability issue.
Many of the supports that have been quote 19 considered unstable can be analyzed; you just have to use a 20 different technique.
21 MR. SHAO: Large deformation.
MR. SISKIN:
Not largo deformation; many of 23 l
them are used as part of the model of the piping system; 24 not try to analyze it as an individual item.
s 57 1
But, again, that's something I would defer 2
until we can address the whole stability question.
3 As you requested I'll be here tomorrow so 4
if following Mr. Hansel's presentation there are any questions 5
concerning where we interface or something like that, we're 0
prepared to talk about it further at that point.
MR. SHAO:
Does the staff have any more I
questions?
9 (No response.)
10 MR. SHAO:
Maybe I turn to Ms. Case [ sic}.
II
... laughter...
l 12 MR. SHAO:
Ms. Ellis.
Sorry.
13 MS. ELLIS: ~Okay.
There are some questions I4 which I understand are going to be bound into the transcript, 15 which we've already given you.
Of course, we'd like to have 16 some answers to those.
I7 In addition there are a few things that will 18 be due today, just as I was listening, that are of concern.
l One thing is, I understand--and please 20 correct me if I'm wrong--this is not your usual 79-14 walk-21 l
down, as perhaps some other plants where you've done the 22 entire walkdown, from scratch, so to speak.
Is that correct?
23 MR. SISKINs That's correct.
MS. ELLIS:
So, really--
MR. BECK:
Excuse me, Juanita, i
0 i
o 58 1
For the record, I need to interject something
~~
2 at this point, I think, that's appropriate, particularly 3
with regard to the list of questions which you provided to 4
Mr. Noonan.
5 The purpose of this meeting as far as we're 6
concerned is to respond to staff questions and comments as 7
we understand the governing principles that are associated 8
with meetings of this type.
The public is invited to attend 9
and to listen.
10 We do not understand that the public is here 11 as a participant in the meeting.
It's not the purpose of l
12 this meeting to serve as a substitute for discovery pro-13 cedures with respect to matters that are in litigation 14 between TUGCO and CASE, nor, I'm fairly certain, would the l
15 staff intend to set a precedent in a meeting such as this,
~
16" to use as a discovery device for use in litigation.
17 Because of the pending litigation, and as 18 long as there is litigation pending, we're constrained to I'
decline to respond or to submit to that form of interrogatory.
20 And that would go for our witnesses, in the 2I question you just asked.
l 22 Ms. ELLIs okay.
j 23 In that case, would you please consider these j
24 questions that wo havo asked here to be an. informal discovery 25 request which we will' formalize, for whatever good that may I
_ ~
59 1
do, and try to get answers to during the normal discovery 2
process.
3 I did not really intend to ask a question 4
of Mr. Siskin except to just be sure that I understood 5
what he had said before.
I did not mean to get into 0
questions and. answers with him.
7 As I understand the 79-14 effort which has I
been done here, it does not appear to me that it is appropri-I ate to compare that with any 79-14 effort that may have been 10 done by Stone & Webster at other plants.
II certainly there's a 1ot of difference if l
12 you have someone do a 79-14 walkdown and have someone come back and then check it after they have had an opportunity to correct most of the problems.
15 Another thing that is of concern is that in
~
16" listening to this today, it appears that there were, indeed, 17 some things, though they may have been relatively few, 18 which were identified, have been identified already; and 19 to our way of thinking, when you have a situation where 20 you have already had the people at the plant look at it 21 during their normal QAQC process, when you have then had 22 a 79-14 walkdown, which as I understand it is supposed 23 to be a very thorough walkdown to be sure everything is 24 like it is supposed to be according to the drawings, then 25 i
60" 1
when you have someone else come along and find things after 2
all of that, to me, the significance of those particular 3
items which are then found is increased much, much more so 4
than they would have been had they been found initially.
l 5
Another thing that is of concern to us, and I
we mentioned this in some of the things we filed before, 7
that it still seems to be coming out, is that when you go I
out and look at something, we do not feel that it's appropri-I ato to have a completed list of attributes to begin with 10 on the hope that people will remember things.
II It seems backwards to us;'that you should.go l
I out instea'd and have as complete an attribute list as possible 13 to begin with, and then you hope that your people will go 14 ahead and try to add to that and find anything that's 15 unusual.
Another question which I would pose which may never have an answer is that if you know these problems 1
18 exist, and I believe it was stated that they do have an 19 understanding of the technical issues, if you do know these 20 problems exist, why not go ahead and address them before you 21 l
go ahead and do all these other things, and then have to 22 go back later and address them.
23 l
It just seems a backwards way to do thingo.
24 i.
And another area of concern is that it 1
1 25 appears from what's been said that about a hundred, hundred i
F~
61 1
and five, I think, out of three hundred problems are now 2
being pursued using draft procedures that have not even 3
been finalized, and this certainly seems inappropriate.
4 I guess our overall concern with this whole 5
process is that things are going ahead without having the 0
proper procedures, the proper information available to 7
start with.
It just doesn't seem like it's got the right I
foundation and basis to begin with.
9 That it's necessary to stop and look at 10 what's going on right now and get all of your procedures 11 and things and then do this stuff, instead of vice versa.
12 Another thing that we're interested in, 13 and we've asked this before, formally and informally, as 14 to how specifically Stone & Webster, for instance, is getting 15 their understanding of the technical issues; are they 16'
~
~
getting them from the CPRT, are they actually getting the 17 documents from all of the hearings, all of the findings, 18 or are they doing this themselves?
Where is this understandirg 19 coming from?
And where and how are these specific issues 20 identified by CASE and by STIGNA being addressed by Stone 21
& Webster.
22 I guess one further thing is that we assume 23 that we will in due time get a copy of the revised plan, 24 and if possible, we'd like to get an indication of when 25 that will be avniinh1n.
k
'o a
63 1
I think that's about all.
~~~ ~
2 Thank.you for the opportunity.
3 MR. NOONAN; I have a couple of things, if 4
I could, please.
5 Number one, on draft procedures, whenever 6
the Utility is working with draft procedures, Ms. Ellis, 7
they know they're working at risk, and they fully understand I
that.
And we let them do that, NRC lets them do that, with 9
the understanding, that if we don't like the procedures, 10 it could very well result in something that they'd have to 11 go back and redo.
12 So it's really at their own risk.
13 I have one other thing; that is, I was 14 reminded this morning, I forgot to ask you.
Back when we 15 had the Contention 5 Panel [ phonetic], there were some 30' documents, reference documents that you said you were going I7 to give to us to look at.
I'll just ask you whether or II not you--
I forgot to ask you that.
I can go back and 19 reconstruct that fo'r you, but I know there were some 20 reference documents at the Contention 5 Panel, you said 21 you'd give to the panel, and I don't remember, I don't think we ever got them.
23 MS. ELLIS:
I'd have to check back and see 24 what those were.
MR. NOONAN:
Okay, we can go back and
o 63 1
reconstruct.
2 MS. ELLIS:
I think February 7th--
3 MR. NOONAN:
When Mr. Jordan--we sat in 4
the Holiday Inn up there.
I'd like to get those out if I 5
could.
6 MS. ELLIS:
I'll go back and review that.
7 MR. NOONAN:
Okay, thank you.
I MR. SHAO:
Just for your information, 9
tomorrow we intend to talk to John Hansel to find out what 10 he has found related to 79-14.
I understand the project.
11 has done the 79-14 maybe a year ago, three years ago.
12 And John Hansel has done an adequacy 13 program related to 79-14.
We'd just like to find out what 14 his finding in this area.
15 I think that's the same concern Ms. Ellis 16 '~
has, too.
I MS. EARLY:
Larry, do you have anything more II to say?
II MR. SHAO:
No.
20 MS. EARLY:
Okay.
Then I think that we have 21 concluded for this afternoon, 22 We'll pick up tomorrow at 8 o' clock in the morning, to begin discussing the sampling issue,
[Whereupon, at 3:30.p.m., the' meeting was 25 adjourned until'8:00.a.m., October 3, 1985.]
1 64 1
PROCEEDINGS 2
October 3, 1985, 8:00 a.m.
3 THE COURT REPORTER:
Due to technical diffi-4 culties with my primary equipment, the first few minutes of 5
the meeting are lost.
Since there was no PA system, my 6
small backup unit was not strong enough to pick up the sound.
7 We are beginning, fortunately, with Mr.
8 Hansel's presentation.
9 MR. HANSEL:
My name is John Hansel with the 10 ERC, and I am the Program Director for the Adequacy of 11 Construction and Adequacy of the OAOC Program.
12 VII.C, as probably most of you know, is a 13 self-initiated program which does involve the plant being 14 broken up into various populations for ease of inspection.
15 You take a plant, you need to somehow 16' determine the best way to get at that plant in terms of 17 how can you evaluate it.
18 Our charter was basically two things; one 19 is to determine the adequacy of the construction; and two, 20 to determine the adequacy of the OAOC program.
21 Exphasizing, or relying most on the construc-tion aspect, the best way to evaluate that is through work 23 process.
24 So we're out to collect information on the 25
F 65 1
adequacy of construction work activity, and using statistical 2
techniques to get through that.
i 3
Methodology:
We're going to categorize 4
installed safety related hardware into populations made up 5
of homogeneous work activities or work processes from which 6
we can then draw samples.
7 I'm going to talk you through today with I
samples and show you how we built those populations, and 9
hope that that will explain the methodology and we can 10 understand five of them; then if you want to come look at 11 the remainder, you can do that.
12 But I think when I finish this morning, you'll 13 have a good idea of how we built the populations.
14 MR. SAFFELL:
John?
15 MR. HANSEL:
Yes?
16'
~
MR. SAFFELL:
Do you want to take questions 17 on an as-you-go basis?
II MR. HANSEL:
I would prefer to go through 19 and then come back, because I think to be interrupted with 20 questions, we may never get there.
So I think when I've 21 finished a lot of questions will have been answered.
22 I~would prefer, if I see a discreet point to 23 stop for questions, I'll do that, but I'd almost rather 24 go through it and then deal with questions.
25 You can take any plant and break it up into w
f o
66 I
these basic disciplines.
I think most plants are done that 2
way, where engineering groups are normally organized in 3
civil / structural, electrical and mechanical, which is a 4
good starting point; we did the same thing.
5 From that point forward, then, we added some 6
general categories of work activities or work processes 7
below each of those.
8 Under civil / structural, we put in concrete 9
placement, structural steel installation, hanger and 10 support installations, liner installations, and fill and 11 backfill placement.
12 Again, nothing new or amazing about that.
13 Pretty standard.
14 In the electrical area, we start with conduit 15 and cable tray installation, cable installation, electrical 10 equipment installation, instrumentation equipment installation 17 In the mechanical area, we started out with 18 piping installation, HVAC, mechanical equipment installation 19 and field fabricated tank.
20 with that as a starting point, what you 21 see in the top boxes basically are the same poists that you 22 saw in the previous three slides, I apologize for the 23 drapes, but you have the handout in front of you.
24 We think these general kinds of work processes 25 broken down into three categories.
We then devise for oursn1vs s
=
.EW 67 3
a series of questions or checkpoints that we want to ask urselves to try to get to something that was reasonably
^
2 3
homogeneous in terms of how do you get through the work 4
process and how can you best sample that.
5 We gathered all the drawings and specifications 6,
and the construction procedures, the inspection procedures,
~7 the codes a'nd standards that were associated with this 3DfJplant.ior those kinds of work.
9 It allowed us to start to get familiar with 10 Gi'bbs &. Hill drawings, specs, CPI; we could then start to 11 gain more familiarity and insight into that plant in terms 12 of work processes of what the actual requirements were in 13 terms of engineering requirements.
14 From that, then, we started to analyze the 15 work processes that were used, the drawings and the specifi-16 cations, codes and standards, the attributes and acceptance 17 criteria for those.
Organization, the crafts, which is a 18 very important part; we did not want to fluster into the
)
19 same population the work done by two groups or two companies, sJ 20 or two different inspection groups.
~
21 So we developed ourselves a series of questions
.O 22 that I'll show you an example; it allows us to take all
., 23 this information and start to draw a finer tune and put 24 it into groupings.
25 From that we came out with a preliminary 0
68 1
listing of some popul'ations.
Back when we first started 2
this I think that number was somewhere in the neighborhood 3
of 27 to 30, that we started out with a preliminary study.
4 I'm going to walk you through some examples 5
and let you see.
I think it best that we just get right 6
to the meat of it and give you some examples that will allow
~7 you to better understand that.
I In those examples I'm going to talk about I
a population as it stands today.
I'm going to talk about 10 the work processes that we feel go into that population II and then the attributes associated with each of those work 12 processes.
I3 Then when I get through the examples, I think 14 you'll certainly have a much better viewpoint.
15 During that process of refining down in the-I0 population, we went through these kinds of questions and 17 subprocesses.
18 We said are the work processes the same or I'
very closely the same; are they identical.
Very, very 20 similar.
We asked ourselves that, and that was Cut No.
1.
21 Cut No.
2, then, for each of those work 22 processes, we started looking at the attributes and the 23 acceptance criteria.
Are they the same or do we have major 24 diff,erences.
.25' The same thing in terms of codes and standards.
69 e
1 We tried to keep those--in fact, I believe we have kept 2
those to where.you don't have codes and standards inter-3 mingled with our population.
It's nice and clean.
4 Then we.have one orientation or multiple 5
orientations of each of those work processes.
6 Same thing with crafts; and the same thing 7
with the inspection group.
8 We went through that screening criteria on 9
each population.
10 Now if we take--for instance in the electrical 11 area, we have five major populations at this point in time.
12 I see no reason for that to change.
13 In the electrical area in general, equipment, 14 cable tray, cables, conduit and instrumentation equipment.
15 I'm going to talk at this point about the 16 cables.
17 When you install a cable you go through 18 three basic work processes: prepulling activities, pulling 19 activities, and the termination activity.
20 To talk about the attributes that we would 21 then inspect and check for on each of those work activities.
22 In.the case of prepulling, you're selecting the cable, size, 23,
. type, verifying the color (this is a requirement).
You're 24 looking for any defects that may have occurred, or may have 25 been encountered to date, on that roll or on that pull or
70 1
on that particular cable that comes out of the top.
That's 2
a documentation review, when we're looking at prior defects.
3 We're looking at the raceway, where the 4
cable will be installed in; is it clear, is it free of 5
debri, is it free of anything that might cause damage to 6
the cable when you pull it.
We had to rely upon documentatior 7
review for that.
8 When you get into the pulling, then, you 9
now have the cable, you've inspected the raceway, you've 10 got the right pipe size, color, and so forth, and you're 11 ready to pull it.
12 We're checking to see if a lubricant was used; 13 that's a documentation review.
We're looking and verifying 14 the routing of that cable.
Which raceways did it go through, 15 did it go through the proper raceways; are the e'nd points 16 proper (we verify that).
17 We look at bend radius; that is both 18 documentation review and visual inspection.
19 We're looking at spacing to verify spacing; 20 we're verifying slack, one of the requirements.
21 We're looking for records of the pull tension 22 that was applied during the pull.
We're verifying tl.at the 23 separation requirements have been satisfied.
We're looking 24 at damage which is both visual and documentation reviewed, 25 and any repairs that were made on that particular cable.
71 1
So on the pulling, there are a number of 2
attributes that we will be looking at in that process.
3 In the termination, we now have the cable 4
in the raceway; the end points are there.
5 We then verify the identification.
Show 6
if the testing was conducted, which is documentation review.
7 We also have to do some testing of the I
cable as they are installed today.
I We verify the insulation, if it's proper; 10 that's both documentation review and inspection.
11 We're inspecting the conductors, both 12 documentation review and inspection.
13 The terminals, hardware that was utilized; 14 is the cable properly landed.
Looking at application of 15 heat shrink, both from record and visual inspection, and 16 the securing of that cable.
17 MR. MARINOS:
John, I have a question before 18 you go on; I'm sorry.
19 On the table of terminations, apparently--
20 my perception is that you have resolved the differences 21 on inspection of different cables and decided it unimportant 22 to separate the homogenity into separate populations; 23 namely, large cables, high-voltage cables, would have to 24 go through a certain testing which is not necessarily 25 e
72 1
required for a small-size cable for instrumentation.
2 How did you resolve that difference in order 3
to lump them all together in one population?
4 MR. HANSEL:
The differences for that type of 5
testing are called out in the quality instruction to the
'6 inspectors.
So for that early verification, it was done 7
by documentation review, and those requirements are in I
our quality instructions.
MR. MARINOS:
But there is a reasonable 10 chance that you are not going to select, particularly since 11
,the number of the small cables way overweigh the number of 12 high-voltage cables, and chances are greatly that will 13 not select that cable, how are you going to reconcile that?
14 MR. HANSEL:
Let me go through the next 15 chart and, hopefully, I will have answered that.
16 MR. MARINOS:
All right.
17 MR. HANSEL:
As I indicated in our review 18 process, we went through a number of questions trying to 19 satisfy ourselves that what we had was reasonably homogeneous 20 and would apply to all sizes of cables.
21 In reviewing those three work processes and 22 then looking at the attributes and acceptance criteria, 23 the attributes and acceptance criteria that we used for 24 the cables, all types, all the way from a thermocouple wire 25 up to a larce cable, newer cable. the metrihntee wovo enmmnn
33 1
.I'll talk about the procedures in a second.
2 They came from a common specification, 3
drawings and procedures.
We did not have multiple procedures 4
or drawings or specification's.
5 MR. MARINOS:
What do you mean by common 6~
attributes?
I fail to understand.
Can you give me an 7
example of a thermocouple cable and a power cable, and 8
what are the common attributes that you have decided on?
9 MR. HANSEL:
These attributes, on the 10 previous chart, apply to each and every cable.
11 MR. MARINOS:
Indeed.
12 Now if we go to the testing, could you 13 expand.
What kind of testing are you talking about of one 14 cable and another?
I mean the power cable--
15 MR. HANSEL:
I'll ask Burt Shair who is our,
16 lead electrical engineer to address that.
17 MR, SHAIR:
Badically, the testing we're II doing is signal tracing at this point to verify routing I'
of cables.
That's what's intended by that particular 20 asterisk.
MR. MARINOS:
What about testing of high--
22 that you do for high-voltage cables that you would not do 23 for--
MR. SHAIP.:
We're,not doing any high plotting 25 of cables.
~
...----.,--e
.m
i
)
74 1
MR. MARINOS:
You mean at the plant you do 2
not do, or as--
3 MR. SHAIR:
The plant has done high prods, 4
but as part of our program, which is basically construction 5
adequacy, we are not retesting cables.
6 MR. MARINOS:
How do you establish the 7
integrity of the cable, then?
8 MR. SHAIR:
We physically look at it for 9
the attributes that are indicated.
Also we do document 10 reviews on those that are not--
11 MR. MARINOS:
You cannot tell when a cable 12 has been damaged, the insulation of a high-voltage cable 13 has been damaged--a medium size cable--I don't know if 14 it's high-voltage.
15 MR. SHAIR:
We're concerned with the construc-
-16 tion process and was the plant constructed, we cut off after 17 installation--there is a high plot done on that cable.
We 18 look at that documentation.
19 At that point it's acceptable from the 20 construction adequacy point of view.
If a problem occurs 21 later that becomes under maintenance function which the 22 Utility will catch in their normal testing program.
23 That's beyond the scope of what we're looking 24 at.
25
a 79 i
1 MR. HANSEL:
I think that's a very good point.
2 We looked at the completion of construction 3
and acceptance by quality.
We have not reviewed the hardware 4
from the point of completing the construction to where it 5
is today.
6 MR. CALVO:
But you plan to establish the
. 7 integrity of the installation that you are checking, the I
integrity of that installation?
I MR. HANSEL:
Yes.
10 MR. CALVO:
And the integrity of that cable 11 installation is not going to be established if that cable 12 is not selected for review by you, particularly the medium 13 size cable.
^14 MR. SHAIR:
As it turns out our distribution 15 is fairly representative, we do have one six-point non-power 16 cable, for example.
We also have. thermocouple cable, -
17
'six power cables.
18 As you would surmise from a random sampling, 19 the majority of cables in the plant that control the instrumentation, and we do have a predominance of those kinds 21 of cables.
22 However, we do have power cable represented.
23 It appears to be representative of the sampling process.
24 MR. CALVO:
But you're not biasing your 25 sampling in order to include one of those cables?
'E 1
MR. HANSEL:
No, we did not.
2 We randomly, in this particular case--
3 MR. CALVO:
And the ratio I would say is 4
10,000 to 1.
5 MR. SHAIR:
Well, we've got one 6.9 KV power 6
cable.
7 MR. CALVO:
You have selected--
I MR. SHAIR:
It happened to fall out.
9 We also have a thermocouple cable on the 10 other end.
11 MR. CALVO:
I'm sure you'll answer this 12 question as you go along, but let me summarize quickly.
13 You mix all the cables together and I just 14 want to say, and you can deal with it.
15 You took a 6.9 KV cable, a fairly complicated 16 termination; you know, the wrapping around with all these 17 things, the installation and [ inaudible.)
18 And then you've' got a thermocouple; all you do, you take a piece of wire, you wrap around the screw 20 there, and the part that bothers me, all of a sudden those 21 two entirely different, they're homogeneous.
That's 22 what bothers me.
23 MR. HANSEL:
Again, we're looking at work 24 processes, realizing you have a number of different kinds 25 of cables, all the way from thermocouple wire, of a very
77 1
small gauge, to a 6.9 KV.
But the work processes associated 2
with those are the same.
3 MR. CALVO:
I hope you tell us about that, 4
because that is not coming through, John.
Maybe later as 5
you go along.
6 MR. HANSEL:
Yes.
7 MR. CALVO:
Okay.
8 MR. HANSEL:
Again, the requirements for the 9
installation of those cables came from a common specification 10 drawing procedure.
11 There is one pulling procedure which is a 12 Brown & Root procedure for the pulling of all cables, 13 regardless of size.
I4 There is one termination procedure ~for all 15 cables and that's Class lE and non-Class lE.
16 The lighting cables and NIS cables, we have 17 this Brown & Root procedure, and for the NIS we have Brown 18
& Root procedure EEI-18.
So we did not have a lot of procedures to 20 work with; there were very few.
21 MR. CALVO:
John, for clarification; from 22 the standpoint of terminations, you have the same procedure 23 that you use for the large cables or the small cables, 24 the same procedure for all of them?
25 MR. HANSEL:
Yes.
N
TR 1
Codes and standards:
Requirements of applicabl e 2
code and standards went into one Gibbs & Hill erection 3
specification 2323-ES-100, and the installation procedures 4
came from that Gibbs & Hill procedure.
5 Brown & Root did all cable pulling and 6
all termination.
So we have one orientation for all the 7
cable installation.
I Brown & Root employees and trained electricians I
did all pulling and terminating.
10 The same crew pulled all cables and same 11 crew terminated all cables.
12 Inspection: TUGCO Inspection procedure that 13 you see noted here is used for cable pulling verification.
14 And in this procedure for cable termination.
15 TUGCO inspectors certified to those procedures and inspected 16 all those cables.
17 Except for lighting and NIS, we had these 18 procedures associated with those.
19-Now it was through that process that we 20 felt that we had homogeneous work processing, yet they're 21 a different kind of equipment, different sizes; but the 22 process of selecting the cable, the right size, right color, 23 the stripping, the inspection of the conductor, the lugging, 24 the termination, are all basically the same.
25
79 1
The only one difference that I can see you
~~
2 could get into is like the twist around the--
3 MR. CALVO:
That's a very important difference, 4
John, between a high-voltage cable and an instrumentation 5
cable; it's a very special process that has to be applied 6
to terminat'e a high-voltage cable.
7 MR. HANSEL:
Right.
8 MR. CALVO:
Special procedures, and a fellow 9
has to be well equipped to do it.
10 Now another guy can--
11 MR. HANSEL:
Our feeling, and I firmly believe 12 that the people who did the pulling and the termination, 13 both the craft and the inspectors, were trained for the most 14 complex, the 6.9 KV, which is probably your most difficult 15 job, all the way down to the thermocouple wire.
Work processen 16 ~
~
basically are the same.
About the only divergence from 17 that is the types we do not put some type of a lug on the 18 end, which you are terminating through a ground screw or 19 whatever.
20 If they're trained for the most complex, 21 they certainly should be able to handle that thermocouple 22 wire.
MR..MARINOS:
How do you establish that this 24 fellow is--he was trained for the complex one, and conse-25 quently he is cable of....
Yes, if that is the case; if
/
80 1
you can establish that, nobody will argue with it.
y Can you make that point?
3 MR. HANSEL:
The same people that are certified to those procedures.
5 MR. CALVO:
John, I'm looking at a cable popu-6 lation.
I got 14,000 cables; let's--for a stipulating 7
discussion.
I've got three kinds of cables I'm going to O
pick, pick up the high-voltage cables.
I'm going to say 9
that participation, have 10 percent of that population.
10 I'm going to say that about 98 percent--no, 11 about 90 percent, 98 percent--10 percent for the high-voltage 12 cables, and I'm going to pick up something like about 88 percent for the medium size cable, 120 volts to 125 volts.
s 14 And the other two percent is for the small 15 cable, thermocouple, that kind of stuff.
16 So by the fact that we--we go to the 17 assumption that the cable population is homogeneous and we 18 go back to this 95-5, and we select a random of 60 items,
- 19 it's a very high probability that very few of those two 20 percent of the small cables, very few of the high-voltage 21 cables will come up in those 60, and the majority of the 22 items selected in those 60 items will be the medium size 23 cables.
24 So on that basis, the possibility could be 25 that you may not even have any thermocouples in there.
81 1
So then you go through this inspection, you 2
verify those 60', could be none of the medium size, and you 3
conclude because a good job was done on those 60 all the 4
others might be okay because it was not considered in my 5
sample.
6 If it was not considered, maybe one percent, 7
maybe one or two that you only look at.
I I'm having trouble rationalizing that in 9
my mind, how can I say, based on that all the 14,000 cables 10 in this plant, they are okay?
11 Even though some of those are not even included 12 into the sampling of the 60.
MR. HANSEL:
Again, we have common work 14 processes, common people, common procedures, common specs.
15 MR. CALVO:
But you've got people from the 16.
time the plant was built ten years; you've got common people 17 for ten years the plant has been bui!.t, because you feel 18 the common people are attributes in the work process.
19 Then you've got common people from the time they start 20 building the plant to now.
21 That's the kind of thing, I hope we can 22 go back to that one later, but this thing with the cable 23 that I'm having problems trying to understand what you're 24 i
trying to do, so we can in some kind of way get this popu-i lation--
~
82 1
MR. HANSEL:
And this sample for this 2
population we did get some of every type of cable.
3 And the distribution of that sample was about 4
as I would have expected; it did include some of every one.
5 Maybe not enough in your mind--
6 MR. SHAIR:
That's right.
Of course, one 7
of the fringe benefits as you explained, the density of the 8
construction of cables are going to give you more control 9
and instrumentation, as it turns out that's the way it is.
10 By doing that, of course, you're talking 11 multiple conductor.
We get much more than 60 terminations 12 out of our population.
I don't know what the total is but 13 it's substantially more than 60, so we sampled a representative 14 number of terminations.
15 MR. CALVO:
Yeah, but--true.
But--
16' MR. SHAIR:
And based on what we're saying-17 about the common craft which is grounded with electrical 18 construction, common procedures, if we find that what we've 19 sampled is acceptable, based on the statistics we're going 20 to extrapolate that to say something about the rest of the 21 cable.
22 MR. CALVO:
Yes, but how can you--
You have 23 to prove that the same craft that did the 6.9 KV was the 24 same craft that did the medium size cables--
MR. SHAIR:
It is.
-n-vr
83 1
MR. CALVO:
How do I go to verify that's the 2
case?
For the last--since the plant has been built.
3 MR. SHAIR:
Well, it's always been Brown &
4 Root Construction; we know that.
5 MR. CALVO:
I know, but at the same--the same 6
people-in the draft, totally concentrated with the high-7 voltage cable; nothing in that?
All the same people that 8
concentrated with the thermocouples?
9 MR. SHAIR:
Well, we know, also, that there 10 are crews dedicated to terminations.
Now that doesn't 11 mean to say that that crew doesn't do everything from 12 thermocouples to power cables.
There may be specialists 13 for certain special activities.
14 MR. CALVO:
Yeah, but if they make a special 15 activity, that special activity by itself made a population 30 by himself?
Because you're defeating your purpose.
MR. HANSEL:
No; we don't believe so.
IO I think there's another key point that's--
19 two points I'd like to make.
I understand where you're coming 20 from.
The first one, as Burt indicated, we're not 22 going to_look at just 60 terminations.
Our random sampling 23 will include 60 cables; there will be a number of terminations 24 associated with each of those cables.
What.that number is 25 I don't know right now.
.m.
84
'l I don't know right now.
2 But we will be looking at a lot of hardware.
3 So you have that sample, plus the engineer's sampling on top c
4 of that that will also give us some additional samples.
5 Another point; when.we address at the completiom 0 -
of the program, when we address the adequacy of construction 7
of-cables, for instance, I will be-addressing what we received I
out of the hardware inspection program and the documentation
^
9 review program.
10 we will also be addressing our knowledge of 11 the drawings and specifications and the construction pro-V 12
.cedures and the process controls.
What we may know at that 13 point in time about any margins or safety functions that are I4 involved in those particular populations.
15 So this is one piece of input to us.
It's a 16'
~
good input.
It does talk to us and it does address the J
17 work processes.
I8 Being an electrical. engineer, I could well I'
. understand your concern between a 6.9 KV cable versus a thermocouple.
21 We have looked at this sufficiently to answer 22 our own minds that the work processes are the same; no need to go back through all the other common attributes and commonality procedures and specs.
25 But we do have, I believe, a good reason to i
85 I
go through and put those into one population, based upon our 2
review of those drawings and specs.
3 Why don't we_go through some other populations 4
then we can come back and revisit this in general, if ycu 5
like.
6 MR. SAFFELL:
Let me offer one other item' 7
which I think is of generic interest to me, anyway.
I
.You mentioned that the procedures employed I
were the same and that training was provided.
I guess I 10 would think it would be key to demonstrating the homogeneity 11 of the population to show that those procedures and/or training 12 did not undergo significant change during that process.
13 MR. HANSEL:
I think, Burt, you have looked 14 at these procedures, and we have a list of them.
Let's just 15 talk about the craft procedures for a second.
16 MR. SHAIR:
Just very quickly; you're right.
17 They have gone through numbers of revisions.
Some of these 18 have been revised multiple times.
19 Our evaluation basically was to see if the 20 attributes had changed, and what we're concerned about 21 is the attributes.
22 And although the language changes and the 23 packaging changes (and it has changed from time to time),
24 the fundamental attributes are the same.
25 And based on that we're saying that the l
86 1
procedures have not changed.
2 The words have but the fundamental attributes 3
have not.
4 MR. CALVO:
I don't care whether the attributes 5
have changed or not.
I don't care whether the quality control 6
the inspection, did a good job or not.
That's why you are 7
here; to demonstrate that the self-initiated adequacy I
construction is going to tell me I have quality in this 9
plant, okay?
10 So I have no problems with procedures changing 11 or revolving.
12 But I only have trouble with what kind of 13 options you're going to put in there.
But I still have the 14 homogeneity based on the commonality, based on the work 15 process.
It's based, the process that has been the same 16 from square one up to now.
17 And you must show that the same people who II did the six, one item in the population, the type, and the 19 same people to do the other one, because when you got 20 specialties doing different things, then you're going to 21 have as many populations as you have specialities.
22 That's the basis of homogeneity that you're 23 using.
Not the attributes, not the procedures, because 24 the end result demonstrates quality, whether it's bad 25 attributes, bad procedures--
F 87 1
Go ahead.
2 I'm focusing on the power that we had problems 3
with them, and we like to go back to those later.
4 MR. HANSEL:
All right.
5 condsit, another population, we broke that into three 6
work processes: the selection of the conduit, the preparation 7
of the conduit and the installation of the conduit.
8 In the selection, you're looking at the 9
size and material as specified on the Gibbs & Hill drawings.
10 The preparation of that is that you're cutting 11 that conduit, the length, you're threading it, you're coating 12 the raw ends, and you're preparing that conduit for instal-13 lation.
You're deburring, any repairs that are involved in 14 the prepping of that conduit.
15 Ir the case of installation, again, the 16 identification of the conduit as required by the drawing, 17 where the end points, origin and destination, verifying 18 that, the fittings, any bends that are involved (that's both 19 documentation review and inspection), the pull points for 20 that particular piece of conduit.
You're looking at flexible 21 conduit if it's involved, and how it is installed.
22 The interface between conduit and tray, is 23 it installed properly.
You're looking for clearances; you're 24 looking for fire stops, barriers and seals, and separation 25 of the conduit.
F o
88 1
Again, we went through the same type of 2
review to verify that these work processes were homogeneous.
3 We have basically two procedures involved 4
for all types of conduit, with the work process down the 5
left-hand side.
6 We have Brown & Root procedures 19 and 19A, 7
applies to both rigid and flex.
Everything was done by I
Brown & Root and inspected by Brown & Root.
9 The same two procedures apply throughout, 10 with a'diffe~ rent procedure for fire barrier.
11 So, again, we have common procedures, common
-12 requirements, common craft, common inspection, common 13 processes.
That one is nice and clean and I don't see 15 any major problems with that.
Concrete placements:
We broke this into 17 three work processes: preplacement, placement, and post-18 placement.
19 The attributes in the preplacement area we 20 looked at and a lot of this you can well imagine'is documenta-21 tion review only; but we looked for verification of the 22 design mix, what is called for on the recipe basis, have 23 the design mix verified.
24 We're looking at the installation of the forms, 25 were they called for, were they verified, how were they b
89 1
verified.
2 We're looking at the size, the location and 3
the size of the preplacement, positioning.
We look for 4
varification of the rebars, all the rebar applications.
5 verification of cadwelds.
On embedments, 6
we're doing both the documentation review and reinspection 7
to verify that the embedments were properly placed, in the I
right--
I MR. SHAO:
What do you mean by embedment?
10 MR. HANSEL:
Pardon?
11 MR. SHAO:
What do you mean by embedment?
II MR. HANSEL:
A plate.
It could be a support 13 that is embedded, anchor type, into the concrete.
14 We're looking at the placement of the anchor 15 bolts and the locations of those, the size and type.
I6 Water stops, that's documentation review only.
II And the construction joints were documentation 18 review only; were they properly prepared.
19 In case of placement, we reviewed the 20
. procedures, and that was documentation review only.
Were 21 vibrators used and were they used as required for in the 22 procedure, both in terms of size and vibration, and frequency.
23 That was a documentation review; and concrete temperature 24 records; how were they maintained,, and that was documentation 25 review.
90 1
Postplacement; we're doing visual inspection 2
of the surface and cored holes, cracks, patches that may 3
have been applied.
4 Curing records associated with that which is 5
a documentation review.
6 And we also have done a documentation review 7
of the compressive strength test results.
8 Basically that is our concrete population and 9
the attributes.
10 MR. SHAO:
Before you leave civil structures, 11 in the other table you have structure steel liner, you have 12 review and factory items.
I don't see it in the view graph.
13 MR. HANSEL:
I'm only going to talk five 14 populations today, Larry.
We can talk to any of those you 15 like, we're not prepared to today, but I'm only going to N
16 address five examples today.
17 MR. SHAO:
I see, but eventually you're 18 going to cover structure steel--
19 MR. HANSEL:
Not today.
20 I can; we've got the data at the site but 21 I don't have it here.
We can go thrt. ugh and discuss any one 22 of our populations with you.
23 MR. SHAO:
Okay.
So you're just showing 24 the concrete--
25 MR. HANSEL:
Showing five cases, to try to
91 1
get the idea of how we built the population, and they're 2
based on homogeneous work attributes.
3 Now, in the case of the concrete placements, 4
we had eight Gibbs & Hill concrete specifications and it's 5
the base document for all concrete placements and all pours.
6 All concrete placement was done by Brown &
7 Root.
I I
The procedures and inspection checklist were 9
the same for all pours, whether you're talking of a one-foot 10 cube or something the size of this room.
Procedures and 1
11 checklist are the same.
12 Requirements for testing were all the same.
13 And, of course, all concrete pours are subject 14 to ASTM and ACI codes and standards.
15 So that is the concrete.
Again, it's common I0 procedures, common personnel, same organizations.
I7 MR. SHAO:
[ Completely inaudible to the II reporter.]
I9 MR. HANSEL:
I'm sorry; I didn't hear you.
MR. SHAO:
Appendix P--QAQC No.
11---
I MR. HANSEL:
That's correct.
22 That population, anything we find, of course, we correlate with the other problems done by the TRT.
That data is taken into consideration.
When we close out this population, we will take every piece of data we know from our
92 1
self-initiated program, plus the other issues that have been 2
in research.
3 MR. SHAO:
After you finish the program, 4
you're going to address all the concerns that was--
5 MR. HANSEL:
Yes. Population by population.
6 With that, I'm going to turn to Burt Shair 7
and ask Burt to talk about instrumentation equipment instal-I lation.
9 MR. SHAIR:
I'm Burt Shair with the ERC.
10 Instrumentation Equipment has two work processe s, 11 one dealing with the installation of tubing and components; 12 and the other dealing with the installation of the instrument
(
13 itself.
14 The attributes that we're looking at for tubing 15 and components are material identification; is the correct I6 material installed as per the building material.
Is it the 17 correct size, and does it have the correct marking on S.t.
18 regarding separation and color coding.
I9 Is it routed from the correct to-and-from N
point as shown on the drawings.
21 If there's a slope requirement because of 22 weather process, does it have the correct slope.
23 Is there an air gap for expansion; and is it M
properly separated from its redundant counterparts.
25 We look at the bends in the runs to make sure
93 1
there's no crimping or damage.
2 And we look at the flex hose installation if 3
it's required by the installation detail.
4 All of these attributes are recreatable; 5
they are physically inspected; and our documentation review 6
concentrates on the qualification of the inspector and the 7
procedures to which he inspected.
8 The instrumentation installation itself is 9
similar--
10 MR.~ MARINOS:
Burt, can I ask a question on 11 the previous area?
12 MR. SEAIR:
Sure.
13 MR. MARINOS:
The complements; I can understand 14 all those attributes you're looking at for tubing, piping, 15 but componentwise, could you give us an example; what is it 16' that you're talking about?
17 MR. SHAIR:
Fittings.
Basically what we're 18 looking at is an installation detail.
Every instrument 19 is referred to an installation detail that shows the tubing 20 run, it shows any fittings that may be involved in that run,-
21 MR. MARINOS; Okay; you're not including in 22 the components trans-- [ inaudible) and things like this that 23 are instrumentation equipment?
24 MR. SHAIR:
No.
25 we start this population into the route valve e
94 1
out to the instrument.
In-line components are in the piping 2
configuration populations.
We're just looking for route 3
valve out--we're looking at anything that's beyond the route 4
valve.
5 MR. MARINOS:
What happens with the component 6
such as the transdeuces that--
7 MR. SHAIR:
If we have transdeuces that have I
electrical interfaces, we pick the electrical interfaces 9
up in the cable population.
We're not looking at cable--
10 MR. MARINOS:
You're not looking at the 11 instrument device itself?
12 MR. SHAIR:
That's right.
I3 Again, the construction is taking an instrument 14 or tubing instrument and they are installing it.
15 gg,re looking at the installation of that I0 instrument and its associated tubing, physically, basically 17 physically.
MR. CALVO:
Well, what do you establish for I'
the instrument--what are you trying to establish?
What kind 20 of integrity are you trying to establish on the instrument, I
what is there, or what looks the right size or....
MR. SHAIR:
As we show, we're looking at the ID.
If we're talking an instrument that has an ID and a tag, we look to see if the tag is the proper one that's there, 25 we can see it, identify the instrument.
95 1
We look to see that it's installed properly, 2
in the right location, per the drawing.
Instruments are i
3 identified on location drawings.
,4 We look to see that it's separated from 5
redundant instrumentation, redundant systems, and we look 6
for damage, such as external damage, anything that may be 7
obvious.
8 MR. CALVO:
The identification you're talking 9
about is whether there's a proper nameplate on that instru-10 ment--
(,11 MR. HANSEL:
Part number--
11 MR. SHAIR:
If the nameplate is the right
~ 13 nameplate.
14 MR. HANSEL:
Part number, model number and 15 serial number if that's on--
16 '
~
MR. CALVO:
You're not establishing the 17 integrity of that instrument?
II MR, SHAIR:
No,
s
< l9 MR. CALVO:
How's that going to be done?
20 MR. WESTERMAN:
Tom Westerman with NRC.
21 How are you looking at the pedigree of the 22 material (balance inaudible - no microphones].
23 MR. SHAIR:
In this case, instrumentation is 24 bought,specified item, it's bought vith a part number, it's
, 25 received that way.
We look at the bill of material; we i
L
o 96 I
check physically against the bill of materials.
If it's in 2
conformance with the construction document, which is the 3
bill of material, and everything as shown on the bill of 4
material, we accept it.
5 MR. HANSEL:
We're not going, Tom, back into 6
the records to verify what was done at receiving inspection.
7 We're taking that item as it was received from the warehouse 8
-which came through the receiving inspection process.
We're 9
not going back and verifying pedigree of that through a 10 data package review or anything of that nature.
11 -
Again, we're looking at the construction 12 process, and--
13 MR. MARINOS:
So the QC inspection is not 3
14 part of your quality construction review?
15 MR. HANSEL:
Receiving inspection is not.
16 MR. CALVO:
QC inspection, of course, covers 17 many' things; and one of the things that I'm thinking is 18 establishing the integrity of the instrument that is installed,
19 so that it will perform as expected.
20 And that is something that you're not doing.
21 MR, HANSEL:
We are not looking at the 22 receiving inspection aspect of the program; we're looking 23 at the program from--during the construction phase, not 24 beyond point of receipt..
25 MR. CALVO:
Just one comment; I just want to
97 1
know--I don't need no discussion on it--the elec'trical one, 2
you went around and you looked at the documentation problems 3
and reviewed the documentation.
4 Some of the instrumentation you didn't bother 5
to go and look at the documentation.
6 MR. HANSEL:
We do look at documentation.
7 MR. CALVO:
All right; there was no notes I
in this particular population as relates to what you did I
in the electrical line.
10 MR. SHAIR:
Yeah, that's right.
11 We didn't make any distinction on this new 12 graph because in this particular case these attributes are 13 all recreatable; there's none that are not.
14 But in the document review we go and look 15 at the inspection packages of these instruments and we 16 do verify that there are documentations there that ar'e 17 spelled out that the inspector was qualified to the procedure, 18 to do.the insp6ction.
19 MR. CALVO:
Thanks.
20 MR. SHAIR:
We'll get into this on the next 21 slide.
MR. BECK:
John Beck.
Burt, if I may, orie 23 of your questions had to do with instrumentation function-24 ability as I heard it, and that's part'of'the testing program.
l 25 It comes under another ISAP, but not part of the construction l
98
-1 adequacy investigations that we're talking about here today.
2 It is being covered but not by this particular 3
program.
4 MR. COUNSIL:
Bill Counsil, TUGCO.
One 5
other aspect that is not being covered in this is the actual 6
design of the instrument, the performance function.
That 7
falls under design adequacy, part of the system.
O MR. MARINOS:
I understand that.
9 MR. COUNSIL:
In addition to that, those 10 instruments, any systems that you presently have, the tests 11
~
on this plant, those instruments have been thoroughly checked 12 out at operating temperatures and pressures.
13 That also is available for review.
I There is also a checklist in the plant today, 15 a limited number of instruments that have been replaced 16 through channels, and so forth.
Those instruments are on 17 a checklist, and the next time, again, rechecked as a 18 functional check at operating temperatures and pressures.
MR. MARINOS:
Well, John Hansel's earlier 20 statement was that by this inspection they will establish 21 the integrity of the installation, and to my understanding 22 the integrity of the installation includes this kind of 23 testing.
24 MR. COUNSIL:
We had to draw a line somewhere.
25 MR. SHAIR:
Maybe we'll have to go through l
i
99 1
some of these--the five questions that we asked ourselves.
2 The installation work process is the same 3
for all instruments, and the procedures and specifications 4
that we referenced here are the procedures that are invoked.
5 Brown & Root construction procedure, ICP4, 6
applies to all instrumentation on this plant.
7 There are two engineering specifications 8
that govern, 2323-ES-100, which is the electrical installation 9
spec.
It deals with separation, requirements.
10 The Gibbs & Hill spec MS-625 deals with the 9
11 major part of the subject, tubing and fittings.
12 TUGCO's procedure 11.8.5 is the procedure I
that governs the inspection of these systems, common to 14 all instrumentation.
15 The instrumentation group of Brown & Root is 16' responsible for the installation of all instruments; and the installation was done by Brown & Root instrumentation 18 craftsmen.
II The inspection was done by TUGCO inspectors 20 in accordance with the above QI.
21 MR. CALVO:
A question--you have air; you 22 also got electrical instrumentation.
All those are combined 23 in the same population, correct?
24 MR. SHAIR:
As far as the devices themselves 25 are concerned, fes.
~_
100 l
MR. HANSEL; I think it's worth clarifying 2
that the electrical wiring to instruments was picked up 3
in the cable population; again, because of that process.
4 With that, I'd like to have Ron Tate come 5
forward and talk about--
6 MR. CALVO:
That's okay; you only got one.
7 THE REPORTER:
Ms. Early, I'm going to have 8
to move these microphones.
There is so much interference, 9
such as coughing, other small noises, knocking out words.
10 MS. EARLY:
Let's take a short recess.
11
[Short recess.]
12 MS. EARLY:
I think we are ready to resume.
13 We broke to move the microphones and for a short recess.
14 You are all going to have to, once again, 15 speak up.
Jose.
17 MR. CALvo:
Yes.
I'd just like to say some-I thing.
At the beginning of the meeting I indicated let's 19 don't bother with the pipe supports.
I just meant that 20 kiddingly.
21
... laughter...
22 I think that my concern with pipe supports, 23 they got satisfactory results, but all I'm trying to say 24 is that equal time with electrical and--
25
... laughter...
101 1
I just want to be sure the record understands 2
that.
3 Thanks.
Go ahead, John.
4 MR. HANSEL:
I can understand an electrical 5
engineer not wanting to talk about pipe supports.
6 I think rather than,getting into the last 7
population, I'd like to take a few minutes and talk through 3
our process again and some of the methodology, because I 9
think it's very key that we understand that.
10 We went through some of the same frustrations 11 that you folks are going through in trying to understand 12 this.
13 We all are familiar with certain kinds of 14 hardware or systems, and we think " gee, that would be the 15 best way to sample this plant."
We went through all of 30 those machinations and we came back to work processes.
17 I'd like to talk about how we got to that, 18 and the process in general.
I think it's important that 19 we re-emphasize it.
20 Again, and you have this chart in your handout-21 it was one of the early charts.
22 I indicated that we pulled the material that we could get our hands on from the A&E and from the various 24 construction and inspection. groups.
And, believe me, that was a lot of data.
t
102 1
In our documentation we identified everything 2
that we had reviewed.
As we go through--as we went through 3
this process of assuring ourselves that we had common work 4
processes.
5 We went through this questioning process very 6
methodically to assure ourselves that we didn't have two 7
work groups or dual and different standards and codes.
8 We did end up, we split the tanks and liner 9
population because we had two different constructors.
We 10 split those particular items of hardware; it went into 11 another population.
12 To get back to Angelo's concern, when you 13 look at an instrument, you have piping associated with it.
14 You have electrical terminations and cables associated with 15 it.
You have the instrumentation itself associated with it.
16 The electrical cabling in that particular 17 case fits into the cable population because it's a like 18 process.
It would be very difficult, and I would have a 19 very difficult time if I selected as an example instrumenta-20 tion, I may have to send three kinds of crafts or inspectors 21 out to that item; whereas, if I randomly select instruments 22 and the tubing associated with it I can send one inspector.
23 If I go after the cable, I can go after the cables with one 24 type of inspector.
25 A very important part of what led us to these
FT 103 1
populations also is how could we best select samples, and 2
the drawing structure and the specification structure, the 3
same questions, the same five questions lead us to that.
4 How can we best select.
5 As an example, I could have taken piping, 6
but in that piping I would have had a large number of welds, 7
balling joints, configuration, orientation, elevation.
I 8
would have had a large number of attributes.
We felt that 9
in that case I could get a much better inspection by breaking 10 that.
11 Again, in the case of pipe supports, we have 12 rigid, we have non-rigid.- We have large bore, we have small 13 bore.
It was how I could best get to a sample of that plant r
14 and inspect it.
15 The bottom lines of all that discussion I I6 just gave you is that the work process was the best way we 17 could get to it, and we think we have a defensible position.
II We did run a check in the case of the electrica l II terminations, the same procedure, the people are trained to 20 the full procedure and they take the luck of the draw when 21 the work gets assigned.
A person may hook up a thermocouple 22 line today; tomorrow or the next five days he may be involved 23 in the fabrication of a 6.9 KV cable.
And next week something else.
25 They're trained to the full work process.
1
104 1
And the inspectors are trained to the full 2
work process.
3 So I think it's very key that we understand 4
that.
It was the best way that we could test that plant.
5 And we've assured ourselves that all work 6
processes are covered in one of our populations.
7 I haven't done a count to see how many work I
processes we're covering, but it's large, as you can see 9
from the five examples that--
I've given you four so far.
10
--that we have a number of various work processes that we're 11 looking at.
I2 For me.to draw conclusions at the end, 13 rather than talk about valves as a population, I'd much I4 rather talk about how that valve got installed, and connected.
15 Then I'm talking about construction activities.
16 Our role was to measure construction.
If 17 you look at the total TUGCO project you have Ard, Livan &
18 Tera (phonetic] looking at design.
We're looking at 19 construction, and TUGCO has in place a very effective test 20 and checkout, turnover process.
21 So you really have the complete system when 22 you put it all together.
23 Yes?
24 MR. NEVSHEMAL:
My name is Nevshemal.
I have 25 a concern, or I guess a lack of understanding with respect
)
]
g.,..,-_
105 1
to the methodology.
Let me try to scope my concern.
2 This is independent of whether or not it's 3
electrical, civil or mechanical, just the methodology by 4
which you created the various populations that you have.
5 As I understand it there are four elements, l
6 or five elements that create homogeneity within a population 1
7-or that are used as the basis of saying this is a population, I
whether it's cable, mechanical, civil structure.
I MR. HANSEL:
May I address that point first--
E
.10 MR. NEVSHEMAL:
Let me just tell you what's 11 in my mind so you can try to fill the holes in or move them 12 around so I understand.
13 Number one is that there is a similar work 14 processes which are--basically three of them--when I look 15 at the work processes that you h' ave listed here I have a pre-I0 installation, or whatever, and I have a during and I have 17
- an after; whether it's prepull, pulling, termination, that I
is really a pre, during and after.
I9 Then I have--do I have similar design 20 documentation, or is it from drawings; is it from specifi-21 cations, or similar specifications, created by similar 22 organizations.
Then I also have acceptance criteria; is 24 it from one procedure, et cetera.
~ '
Then I have organization where all this was m-._
r_
,r,
,,_.,,e m.. -
-.m
106 1
created from one organization.
2 Then I have crafts.
Are the crafts electricall /
3 orientated, mechanic--
In other words, can you get back 4
to that very basic piece of information, or element of your 5
methodology? what are the clear definitions of the elements 6
of homogeneity that is behind every population?
7 MR. HANSEL:
All right.
I've done that on 8
four populations.
9 I want to make--
10 MR. NEVSHEMAL:
I'm not talking about independ-11 ent population; I'm talking about the general methodology 12 that you're following from which then you apply this to I3 all the various aspects of the plant, and then you create 14 a number of populations that you have.
15 MR. CALvo:
We want to know the ground rules, 16 the basis.
How you went about it.
17 Suppose you only had one procedure in this 18 plant; would there only be one population?
Suppose the I'
electrical instead of one procedure, have three procedures?
20 MR. HANSEL:
I think I know where you're at.
21 MR. CALVo:
What are the ground rules?
22 MR. HANSEL:
We concentrated on defining 23 homogeneous work activities, work processes.
24 And we tested each of those with five questions, 25 Now, you have one of those work processes as
107 1
cable terminations.
Now, how can I best get to cable 2
terminations.
I don't know how many terminations there are 3
in that plant, probably--I'll wing a number--a hundred 4
thousand.
I cannot get to the details of each of those 5
terminations easily, but I can get to those through selecting 6
cables because we have a listing of all cables.
So I can 7
then approach that work process by sampling cables, so my 8
population for selection purposes becomes population.
9 MR. CALVO:
Wait--
10 MR. HANSEL:
Let me talk it a bit more.
11 MR. NEVSHEMAL: Wait a minute.
From what you 12 have said, I have the feeling that a population should be 13 terminations.
MR. HANSEL:
.I'm testing the work process.
15 I have a number of kinds of terminations; how f ar dotyn do 16 you want to cut this?
I may have ten different kinds of 17 terminations.
18 Now do you want 10 different populations for each of those?
20 MR. NEVSHEMAL:
I just want to understand the methodology that you followed, independent of discipline, 22 independent of kinds of equipment, so that I can legitimately 23 say yeah, those are the all-encompassing population.
24 MR. HANSEL:
Let me talk through it again, 25 at the possibility of being redundant.
.._._._..____..m-
108 1
You take all work processes and follow--they 2
will include a' plant You take all work processes, they 3
equal a plant or plant equals all work processes.
4 I then went through and found processes that 5
are the same for cables, for cables of any kind; the work 6
processes are the same for cables, be it preparation, pulling, 7
or termination.
8 We then went through to be sure that they 9
passed the test, because I didn't have a different set of 10 codes and standards for one type of wiring versus another.
11 If.you noticed, in the case of cables, I had 12 a lighting specification, but the attributes of the lighting 13 specification are almost identical, if not identical, to 14 those of the Brown & Root specifications.
15 So we went through; did we have like attributes I6 and acceptance criteria.
We went through these tests; and 17 I will not go back through those again.
II I now had work processes that were homogeneous 19 in our minds, in our opinion.
They were alike, and they 0
stood that test.
21 Then you have to ask yourself, how could I 22 best sample; how could I best get to measuring the effective-23 ness of that work process, termination of cables.
I can sample cables and then inspect the cables.
25 Now you can split that any number of ways,
~. - -, -
. - - -. ~
,.m
109 1
to the finest level you want to; it didn't make sense, doesn't 2
make sense.
3 To test that plant, and to evaluate it in 4
terms of its adequacy, I can best evaluate cable terminations 5
by sampling cables.
I will probably look at--I don't know--
0 our' total, between our random sample and engineer's sample 7
on cables, we're looking at 92 total cables.
I And, Burt, can you swing a number out in I
terms of average--
10 MR. SHAIR:
Three hundred, somewhere in that 11 area.
MR. HANSEL:
How many?
I MR. SHAIR:
Three hundred terminations.
MR. HANSEL:
And we will be looking at 15 approximately 300 terminations of cables.
16 It's not easy to get to a complete listing 17 of all terminations.
I'd have to go to almost all systems 18 and extract from a lot of detailed drawings to get a listing 19 of terminations to select on terminations.
20 MR. NEVSHEMAL:
But, John--
21 MR. HANSEL:
I can get to it easily and also randomly through cables.
23 MR. NEVSHEMAL:
Okay, you're very close to my concern in resolving this.
You said that all of--that the plant is m.-..
.r____._.___
110 1
divided up into work processes.
You have a list of those 2
work processes?
3 MR. HANSEL:
Yes.
4 MR. NEVSHEMAL:
And then you looked at each 5
work process and you said there are similar attributes for 6
those individual work processes?
-7 MR. HANSEL:
Yes.
8 MR. NEVSHEMAL:
And so now we have homogeneity 9
within this work process; this work process has the attributes 10 and they are different from work process B which has these 11 attributes, and work process C which has these attributes.
12 MR. HANSEL:
Yes.
13 MR. NEVSHEMAL:
So we have homogeneity 14 within A, B, and C work processes, whatever they are.
You 15 made a list.
16 MR. HANSEL:
Yes.
17 MR. NEVSHEMAL:
Okay, how does that create 18 a component-related population, like cables, like electrical 19 instrumentation or the ones that you have listed?
How did 20 you get from that point to the list of populations that 21 you have with--I believe there are about eight or nine of 22 them?
23 MR. HANSEL:
There are 25 populations.
24 We got to that point by two means--
25 MR. NEVSHEMAL:
Let me say one more thing.
111 1
This whole thing would be more understandable 2
to me if you were to say that the plant can be divided up 3
- into cable, electrical equipment, blah, blah, and this 4.
sort of thing, and they have similar attributes.
5 MR. HANSEL:
We're saying the same thing.
'6
'MR. NEVSHEMAL:
Oh.
Are we?
7 MR. HANSEL:
Yes.
I MR. NEVSHEMAL:
Okay, that's what I wanted.
9 Let's go from there.
10 MR.. HANSEL:
Each plant and its drawing 11 structures, the construction processes are probably all the 12 same from plant to plant.
I3 I don't know of any--you're going to always I4 pour concrete, you're going to always have cables, you're 15 always going to....
The work processes are probably the II -
same for every plant.
I You can take-. Let me build it from the bottom up, that's where we started.
We took it, if you recall, I said all the pertinent information in the electrical 20 area, in the broad realm.first.
21-And then we started the filtering process 22 and linking together to get into work processes.
23 In the electrical area, what are the common 24 work processes?
25 We built that level which is this level right t
112 1
here, pre, pull--
And then came up to there and said I 2
can best sample on this plant due to their records and 3
the drawing structure, I can best get to those for sampling 4
by cables.
c 5
So that became a population.
6 I could have easily started from cables and 7
went down.
I could have worked the reverse process.
8 MR. CALVO:
Or you could have used terminations 9
and that way also to check cables.
10 MR. HANSEL:
Yes, I could have, and that's 11 the bottoms up.
I start with terminations and I can best 12 get there, but to get a termination, before that I had to 13 pull it and before that I had to select it.
I4 So that's--I hope I've answered it, but that's 15 basically the methodology.
I6 That's what I'm here to measure, is the o
I7 offectiveness of the construction.
MR. NEVS11EMAL: I understand that; I'm trying 19 to understand the methodology, how you created the populations 20 that you have.
21 MR. !!ANSEL:
We onded up with a number of 22 work processes, and we then looked for structures; this one 23 is good and straightforward.
24 We then went from there to a population, 25 which would allow un case of sampling.
113 b
1 MR. NEVSHEMAL: So somewhere there is a 2
generic list of work processes.-
3 MR. HANSEL:
Yes.
4 MR. TRAMMELL:
John, I'm Charlie Trammell, 5
NRC.
6 Maybe it would help if you could explain 7
why it is that--getting back to Angelo Marinos' question 8
about the apparent difficulty--or not difficulty--but the 9
trickier job of connecting a 6.9 KV cable to something 10 versus something as straightforward as crimping a lug on 11 the end of a wire and screwing it into a terminal.
I'm 12 having a little trouble seeing why you can through those 13 into the same basket, because if you had taken, for example, 14 the hardest process and said all these electricians are 15 trained to do anything, as the luck-of-the-draw kind of II things, would it be perhaps too severe a test, then, to 17 do a sample of all of 6.9 KV terminations, and then rest 38 comfortably and pass muster on those by saying "well, I'
those are the hardest, obviously if those are okay, all 20 the rest of the electrical terminations are obviously okay,"
21 and be more comfortable with that conclusion than by saying 22 "well, we sampled all electrical terminations, therefore, 23 the 6.9 KVs are okay," when you know in your heart those 24 are a little tougher,to make.
MR. !!ANSEL:
They're tougher to make, but
7__
114 1
the process, again--I hate to keep hammering on it, but we 2
concentrate on the work process.
3 MR. TRAMMELL:
Is it the same process?
4 MR. HANSEL:
One procedure covers it all.
5 And people are trained to do anything--any and 6
everything in that procedure.
They're trained for the most 7
difficult, those that require less requirement, less concen-8 tration, less skills, all the way down to twisting Jose's 9
thermocouple wire around the screw.
10 MR. TRAMMELL:
You're giving equal weight to 11 that sampling; you're going to go out and--
12 MR. HANSEL:
I'm going to talk about that 13 plant, I need to talk about all work processes and cover 14 all hardware.
If I just selected the most difficult and 15 somebody would ask me about the thermocouple wires, or some-16 body would ask me about the power cables, or someone would 17 ask me about lighting cables, I have to talk the total plant.
8 MR. TRAMMELL:
Let me ask you one more then.
19 I soo your point.
Suppose that you do your random samples, 21 which you are doing, and the luck of the draw has it that 22 no 6.9 KV terminations appear, which could easily happen, 23 I guess, by the luck of the draw, would you end up with 24 a little lump in your throat wondering about 6.9 KV termina-25 tions?
115 1
MR. HANSEL:
Again, I tested the work process.
2 MR. TRAMMELL:
Okay, I don't want to dwell on 3
it but--
4 MR. HANSEL:
Engineer to engineer, I would 5
wonder about that; but, nonetheless, I worried about the 6
work process, and the people were trained.
7 There's no reason to be.lieve that they would 8
do any less good in terms of adequacy of work on the 6.9 KV 9
versus the thermocouple.
10 They're trained; they're briefed on that, 11 and I've reviewed those training programs and crafts.
12 MR. CALVO Let me go back to the basis again.
13 Hypothetically speaking, if I only had one 14 procedure, if I constructed that' plant with one procedure--
15 theoretically speaking, okay--then I would only have had one 16 highway of population.
If all the craf tsmen, the mechanics, 17 the electricians would do as well, one with the other, then 18 only one population.
Take 60 items, satisfy that condition, 19 and I had reasonable assurance that everything is okay.
20 Let me go further.
21 suppose I got five disciplines; mechanical, 22 electrical, civil, structural.
Each one of those disciplines, 23 they had one procedure.
Then I would only have five popula-24 tions.
25 Let's pick out one discipline in there and
I 116 1
let's say, for instance, the,cicetrical discipline, the 2
cables that we're getting into.
3 Now you say that the reason I picked up the 4
cables is because I got one procedure to tell us the crafts-5 men who can install cables, whether they're big, small or 6
insignificant, okay, in accordance with that procedure.
7 Now, suppose I had three procedures instead 8
of one, the one for big cables, the one for little cables, 9
and one for middle-size cables and according to your explana-10 tion I would have three populations, right?
11 MR. HANSEL:
If there were significant 12 differences between those three procedures I would have.
13 MR. CALVO Okay.
14 MR. HANSEL:
Or if they were done by different 15 groups, I would have.
16 Or if they had different acceptance criteria 17 I would have.
18 MR. CALVO:
So then for me to understand why 19 you selected the cables, then I must look at that procedure 20 and procedures have sections and chapters, just like an II FCR has 17--they got a lot of sections, okay? some are 22 different than the others.
23 If I go to that, if I determine that the 24 work process used for the 6.9 KV or the work process used 25 for the thermocouple is different in my opinion, you and I
117 l
l 1
have the difference, have to reconcile the difference.
l l
2 And then the condition, you have two populationn, 3
one for 6.9 KV, one for thermocouple.
I can be, going through 4
those things, and I could, hypothetically speaking I can have 5
a population for each section, if I can prove it to you, l
.6 or prove it to ourselves, that yes, those are the homogeneous.
7 How can a cable for 6.9 KV be on a high pot connection with 8
a thermocouple wrapping around a wire, how can that be l
9 homogeneous?
And I say well, what you have done I will not i
l 10 accept the one procedures that's the reason I have two l
i 11 procedures, because even though you've proven an umbrella 12 of one procedure, it's actually two work activities because l
13 you accomplished two things that are totally different from 14 each other.
1 15 So, what I'm saying, for me to understand 16 what you're doing, I've got to go through all that, and 17 whether I agree with your population is based on the basis 18 and the ground rules that you're setting up for you to j
19 establish it.
We have never got to the point to challenge 20 those or to review those and see whether we agree with you-21 or not.
That's the way I'm going to see it.
i l
l 22 Am I missing my point?
Making it too simple?
l f
23 MR. HANSEL:
No, you make a very good point.
24 But we have had the benefit of doing those 25 reviews, looking at those construction procedures, and the
118 1
inspection procedures, and understanding the. makeup of the
'2 people and the groups, how they were assigned.
We've had 3
the benefit of doing that, and we feel that the work processes 4
are the s. ne.
l 5
HR. CALVO:
All right.
6 So if I go to your files, for me to understand 7
what you have done, so I can either accept it, reject it, 8
-or communicate, will I find in your file your evaluation'of 9
how you arrived to this conclusion of the work processes.
18 I must go back to the time period when Comanche Peak first Il got the construction permit, when the first wire was connected l
12 there.
Are you going to show it to me with some reasonable 13 acaurance that, yes, the same craft through all this period I4 of time, the same people in that period of time on the average 15 have been working with 6.9 KV, also the same people--you're IG going to have a record that is going to tell me to make that 17 determination in your files?
II MR. HANSEL:
Yes.
19 I think on a one-on-one basis and going 28 through with our engineers on a population-by-population 21 L
basis--
22 MR. CALVO:
Now if I found that you had not--
l l
23 that I had.this disagreement and I convince you there's some M
problem in there, then we're going to have to renegotiate 25 a new population; agreed?
l
119 1
MR. IIANSEL:
Agreed.
If we agree, and I see 2
a problem there, I wouldn't have a problem with that.
I've 3
looked at it in a great amount of detail, and I feel com-4 fortable with the process.
5 It's not an easy job--
6 MR. CALV0:
To me--
7 MR. IIANSEL' Let me just talk--the kind of 8
documentation you'd be going through, just bear on this for 9
a second, because wo, basically--create a new spec as 10 compared with A&E, but wo have a population description II written for each population; what is in that population, 12 what are the boundarios of that population, what's excluded, I3 and what are the interfaces, all written out.
I4 We have a listing of overy picco of documenta-I' tion that wo have reviewod.
16 We have a list of all the proceduros, the 17 specifications, the drawings, cablo list, et catcra.
Wo 18 havo that.
19 We have a listing of the attributos and whoro 20 they came from, page and paragraph, or drawings for each 21 attribute, what's the sourcos any attributos that wo havo 22 not included wo havo a justification why they're not included.
23 MR. CALVO I havo no problem finding that 24 information.
Wo can go thoro and chock it out, so you toll 25 me tho ntraiqht.
If T no there to vour filen, okav? and I'm
120 1
going to go there and determine the history of the way 2
cables--any population in that plant, I'm going to find 3
that information available to me, because I've got to make 4
a determination whether the same process was used from 5
the time that a population--okay?
6 MR. HANSEL:
It may be an engineer's notes 7
where they have reviewed the process, discussed with Brown I
& Root or TUGCO, but--
9 MR. CALVO:
Are those notes available to 10 the NRC so we can audit?
11 MR. IIANSEL:
We can show you that.
12 MR. TRAMMELL:
John, lot me ask you some I3 basic questions on sampling.
I4 If you go out in a population and you find, 15 I believe it's zero defects out of 60--
16 MR. IIANSEL:
That's true.
17 MR. TRAMMELL:
--you conclude that you've 18 gono far enough and havo 95-95--or I guess botter than 19 95-95 at that point becauso you're prepared to go further, 20 if you find one in a population of sixty, and then if you 21 find ono defect in the population of sixty, you increaso 22 the same size to somothing like a hundred and twenty-ono?
23 MR. IIANSEL:
Ninoty-fivo.
24 MR. TRAMMELL:
Ninety-five.
And if you find 25 nono in the further, then vou stoo?
=
1 MR. HANSEL:
We stop.
2 MR. TRAM!! ELL:
Is that the end of the--if you 3
find one in that remaining thirty-five, do you expand 4
further, or-do you decide that you failed at that point?
5 Let me talk, or have Al Patterson talk 6
through that process for you for just a second.
'7 MR. CALVO I think that would be good.
8 MR. HANSEL:
Yes, but before he does, I want I
to make a point.
I' We have our random sample of 60, and then II agreed to pick a second sample of items required for safe 12
" * ~ " '
shutdown.
13 There's an overlap between the first sample 14 if, in fact, in the first. sample of 60 there are some items 15 required for safe shutdown.
16'
~
So you,end up--I guess our average sample 17 size is around 90 right now.
Total. items that we're inspect-
~
l 38 ing.
II The same expansion is based on the first 60.
1 l
20 And that's our total at-random sampling.
l 21 Let me have Al Patterson talk through the 22 sample expansion.
23 MR. CALvo:
May I suggest that we postpone M
the answer to that question, Charles, to a later time, and l
25
,,'can nail down the work process, so we can go to t'ho next l
\\
i
a 122 1
item.
2 MR. TRAMMELL:
Oh, we're not done yet?
3 MR. CALVO:
No, we're not.
4 MR. MALONSON:
John, Excuse me..
I have no 5
problem with the homogeneity of a population.
We do not 6
have homogeneity of the item within a population; for 7
instance, we have staid lock [ phonetic) terminations; we 8
have wrap terminations, and so forth, clamp terminations.
9 Did anybody test the sample to see whether 10 or not the sample was truly representative?
11 (The speaker was not at the microphone - it II was very difficult to hear.)
I3 MR. HANSEL:
Let me address that.
We analyze each of those.
To talk about
.I I4 15 about program, I went back on three populations on Braidwood.
30 We had a much more severe case on Braidwood.
And we had 17 eight contractors involved in Braidwood, rather than one II constructor.
I' The worst case we had, we had five contractors 20 doing one type of work.
21 We went back after those samples were drawn 22 and after they were analyzed and we looked for distribution.
23 We had a good distribution there, through the random selectior 24 process of work processes, contractors, inspectors.
The 25 process works with the data available.
123
}..
1 It was a good program.
We're looking at 2
these as we go, not because we're required to, because ps 3
we're interested, and we've seen a good distribution.
I
.e,
- '1 4
think the cables was a good example.
I can't address any I
f 5
others right n,ow, but we know what that distribution is.
6
[MR. MALONSON:
Can we just say theoretically, I
7 what would happen if you found that the sample was not
./
Y8
-(
j representative?
.t) 9 MR. HANSEL:
Again, I am sampling the work
,v l'
10 process which is the homogeneous elements, and if it's
(
11 throughhet, then I see no need to do any further expansion.
\\
12 MR.'MALONSCG No matter how the sampling 13, came out,you would regarcl it. as representative?
M MR. IIANSEL:
Yes.
I would not; I would not, 15 becau's,e, again, I'm testing the work process.
~
~
F 16 MR. NEVSHEMAL:
One of your elements of
)
17 homogeneity is the fact that the crafts have been trained 18 to a particular procedure that covers a variety of cables.
19 Is there evidence, have you seen evidence, 20 that the crafts involved in this activity have been trained 21 to that entire procedure?
Is there evidence existing of
> >I
\\
,,)
A,j 'i this trpining?
1
('
. ' j
ds i
MR. IIANSEL:
They are trained to the total
,l i, y, ',
T
'14 procedures.
We have sampled the craft training process th. rough another issue! plan, and we did find records in those 1
y s
^Y
124 I
we looked at.
2 MR. cALVO:
Before we get off the work 3
process, I'd like you, if you can, finish the presentation.
4 You had a couple more slides.
5 MR. HANSEL:
We can skip to another--we can 6
skip large bore pipe consideration, if you like, and I'll 7
go right to the summary.
I MR. CALVO:
Okay, unless somebody else has 9
some questions, some burning questions on the work process--
10 when he finishes the slides we can go back and talk--
11 MR. HALE:
It's not burning, Mr. Hansel, but 12 you've defined pretty well the way in which you established 13 the work process, the homogeneity processes.
14 And the manner in which you look at the 15-attributes for inspection are re-inspected.
What you've 16' covered today, the development of everything up to the 17 point of actual re-inspection.
Why, then, a re-inspection II of a population?
For example, would you--use cables; the II inspector takes one of the inspection procedures which is N
the QI hookup, go out and inspect, for example, a power 21 cable, and identify or make a decision with respect to 10 22 attributes.
23 Take another sample, randomly selected N
sample out of that same population, for example, a thermo-couple, and come back with only one attribute that applied
125 1
to that cable; is there a soft spot in youi methodology L.
4 2
to the approach to the development of the QI which I, guess 3
is the bottom line of this effort?
4 MR. HANSEL:
I don't think so.
You can 5
have two kinds of conditions occur, Cliff.
You may--the 6
difference between a 6.9 KV and a thermocouple, there may 7
be some attributes that apply to a 6.9 KV that's not applied fi 8
to that thermocouple.
Not very many so you could have that 9
case.
10 And I think the engineers who have the final 11 judgment as to whether or not they're satisfied with that 12 inspection, it's up to them to make that conclusion, because 13 they review those packages at the completion of inspection.
14 The other case that you could have, which 15 I would also be concerned about is if, in fact, some piece 16' of that cable was not acceptable for inspection, which would 17 cause me to not have a good inspection of that cable.
Then 38 I may want to go pull er.other sample, and we've done that 19 in a number of cases due to inaccessibility.
20 MR. HALE:
Now I may be incorrect, but I 21 believe-in fact, you had at least one example of the type 22 I just described, where you came back, or one inspector 23 came back from inspection with only one attribute in the QI 24 checklist that was applicable out of, say, ten.
25 MR. HANSEL:
It may have happened; I don't know.
di4T m ;;, % Q n.
1 But that issue, that population is not c1 _ad '.15"C.
t won ' t 2
be closed until my lead engineers are satisfied that they 3
have enough data on that population to talk about how good 4
it was.
If you had a high percentage of the sample where 5
the attributes didn't apply, I'm sure they're going to come 6
talk to me and say we better go back and do something 7
different.
8 It may have happened; I don't know, but, as 9
I say, none of the populations have been closed, only indi-10 vidual packages, and I would only want to be judged on a 11 closed package.
And I don't know if this one's closed and 12 in the records file or not.
13 Ja; is still in process work and still 14 being evaluated by engineering and by ourselves.
15 So I can look at it when I get back to the 16 site and talk to you about it, but I don't consider it a 17 closed issue until the lead engineer passes it to the record 18 center.
Up until that time it's still being reviewed and 19 assessed.
20 They're looking for those kinds of things.
21 MR. NOONAN:
John, I have a question for you.
22 In all the attributes that you're now looking at, ERC is 23 looking at or you and your contractors are looking at, what's 24 the number?
How many are you looking at?
25 MR. HANSEL:
We're probably going to do in l
a 127 1
the area of a half a million inspections or documentation 2
reviews.
This is an estimate, but it's probably very close.
3 You take 25 populations, we're averaging about 4
nine attributes per population, for a total of 225 attributes.
5 There's some sub-attributes like in welding; there are sub-6 attributes--and there's an average of 4.7 subs per each of 7
these measures which gets you to a thousand fifty-seven.
I We're right at 90 samples per population; we're going to 9
be looking at approximately, conducting 474,000 (that's 10 approximately) individual inspections or observations.
That's 11 not a small number.
I Following through the same logic of what we do in documentation review, we're adding another 60 or 65 14 thousand.
So we're somewhere in the neighborhood of a half 15 a million observations of completed work in that plant.
16 We have about a third of these inspections 17 completed now.
18 MR. NOONAN:
Based on your experience, you've 19 been doing this, how does that compare? high, low, average?
20 MR. HANSEL:
I'm sorry, Vince; I couldn't 21 get all that.
22 MR. NOONAN:
I said based on your experience 23 with other work, not Comanche Peak, but other work, how 24 does that compare?
25 MR. HANSEL:
In terms of numbers?
128 i
1 MR. NOONAN:
Yes.
2 MR. HANSEL:
It's very comparable or very 3
close to Braidwood.
4 On Byron we did about 225,000 inspections, 5
216,000 as I recall being an exact number.
6 MR. NOONAN:
That's good; thank you.
7 MR. HANSEL:
It's comparable to Braidwood.
8 And the program is structured almost identical 9
to Braidwood.
10 MR. NEVSHEMAL:
In the mechanical area, can 11 you elaborate a little bit on the work processes that created 12 your populations in the mechanical area?
13 MR. HANSEL:
Al, can you talk about that 14 briefly?
15 MR. PATTERSON:
Would you restate the question, 16 please?
17 MR. NEVSHEMAL:
I would like to have some 18 explanation or understanding of presentation on the work 19 processes that created the population that you have in the 20 mechanical area.
21 MR. HANSEL:
If I could, and I don't want 22 to beg off of the question--I don't know that Al's fully 23 prepared to do that, and I wouldn't want to be judged on 24 that answer today.
25 7.d rather sit down, either give presentations
129 1
on all 25 populations or have people come to the site and 2
look at that.
3 If you do want an answer, it's purely off 4
the top of the head and not finite at this point in time.
5 MR. NEVSHEMAL:
No, no.
I think what I'm 6
asking for is no more than what you did on the other ones.
7 On the cables you had three work processes; on the electrical 8
you had three.
I'm just asking the same thing on--
9 MR. SHAO:
I understand more advanced in 10 the electrical area, they're starting--
11 MR. HANSEL:
No; we have most of them 12 defined, but I don't want Al Patterson to talk off the top 13 of his head because he doesn't know each of those populations 14 with finite detail.
15 MR. SHAO:
You're getting them ready but 16' not ready to talk?
17 MR. HANSEL:
I'm ready to talk if I could 18 go to the site and_get the data.
I don't want to be judged 19 on something that he's talking off the top of his head, and 20 not'from a piece of data that I've looked at or prepared.
21 We can talk in generalities.
22 MR. CALVO:
If I can, you see, you already 23 established 25 populations.
You already established them.
24 MR. HANSEL:
Yes.
25 MR. CArv0:
So if you established them you
130 1
you must have the work processes defined.
2 MR. HANSEL:
Yes'.
3 MR. CALVO:
And that's what he wants to know.
4 You've got them.
We're asking how you--
5 MR. HANSEL:
We can address--we can talk 6
large bore, large bore pipe configuration.
I'm sure we 7
could probably talk--I can talk any of them, but I don't I
want to--I'm not sure that I have all the data and information 9
here with me.
10 MR. NOONAN:
John, I agree with you.
I don't 11 think we ought to put this on the record if you're not sure.
12 MR. HANSEL:
I'd like to get you with my 13 engineers, lead engineers, who have gone through this, and 14 who have put those populations together and have made those 15 judgments, rather than me talking off the top of my head.
16'
~
~
MR. SHAO:
So you're not really ready to I7 talk today.
II MR. HANSEL:
No.
I could,go to the site and II talk to you, but here right now I don't have the right 20 people with me.
21 MR. CALVO:
But information is available in 22 your files?
23-MR. HANSEL:
Yes.
M MR. SHAO:
The same way with civil structures?
MR. HANSEL:
Yes.
131 1
1 MR. MARINOS:
What about electrical equipment?
l
)
2 Are you prepared to give us any detail?
3 MR. HANSEL:
I have the lead engineer here 4
for that; we could address that.
5 MR. CALVO:
But, again, we go back--we still 6
got the fundamental thing that we're going to have to 7
determine, see what they have done, whether we agree or I
disagree with it, and the work processes, based on the I
homogeneity of the procedures that they used, the craft 10 being there for the life of the plant, the same people; we II got to do that.
12 MR. HANSEL:
We came prepared to talk five 13 today, that are somewhat representative.
I didn't want to 14 get into specifi~cs.
I'd prefer not to.
If you like, we 15 can talk as best we can.
16 MR. SHAO:
You're not ready to talk on civil 17 structure or mechanical, but you're ready to talk large bore 18 and piping?
19 MR. HANSEL:
Yes, large bore piping configura-20 tion.
21 MR. SHAO:
Let's hear that.
22 MR. MARINOS:
Specifically what I'd like you 23 to explain to me is how do you lump the same [ inaudible],
24 and what are the attributes you look for when you go through 25 your inspection of those two different components?
132 1
MR. SHAIR:
I can give you a general idea 2
of what we did in electrical equipment if you like.
3 Again, basically, the work process is an 4
installation process.
We're talking about taking a piece 5'
of equipment, whether it be a panel, a motor, or whatever it 6
might be, and installing it.
7 So the kinds of things that we're looking for, I -
again, are similar to what we've already talked about.
We want to verify the identification of a 10 piece of equipment; want to make sure it's correct, and 11 that it's correctly identified.
12 We look at the installation with regard to 13-how that equipment is supported.
We look at it to see if 14 it's been modified or sampled in the field.
If it has been 15 modified or assembled, we look at that modification to make 16 sure it's been done in accordance with either manufacturer's 17 instructions or issues of drawings, or change notices.
18 We look at the location and orientation to make sure that physically it's in the right place, whether 20 entered properly with regard to separation requirements or 21 whatever other requirements dictate the orientation.
22 Again, we're going back to the information 23 presented to the constructor which'is basically, in these 24 cases, the drawing, installation drawing or location drawing.
25 MR. MARINOS:
This kind of a thing, then, is
133 f
1 no different than whatever you'd be looking at.
2 You very well have lumped pumps and motors 3
together, and the transformers and breakers.
4 MR. HANSEL:
All electrical equipment is in 5
one population.
6 MR. MARINOS:. The pump is not, so, you know--
7 MR. SHAIR:
The pump is a mechanical piece of 8
. equipment.
9 IThree people were talking at once, therefore, 10 statements were inaudible.]
11 MR. MARINOS:
It appears to me that the 12 features you look for in the installation are no different 13 than any other mechanical component.
14 MR. SHAIR:
Except we're dealing with electrical 15 equipment.
16' MR. MARINOS:
Yes.
17 MR. SHAIR:
And installed by the electrical 18 crafts, and installed by the electrical work spec which is 19 the ES-100 that we talked about earlier.
20 Everything that we do is based on ES-100, 21 which is the electrical, the one electrical installation 22 work spec, installation specification.
U MR. HANSEL:
And, Angelo, you just hit on it, M
the work process.
D MR. CALVO:
We want to determine what you
134 I
have done.
2 The part--you know, this is a most important 3
thing in your whole program plan, the CPRT, construction 4
adequacy, implications of finaudible) in its populations.
5 If I'm you, I hear myself on the telephone, 6
I will be driving Noonan nuts asking him, "Why you guys 7
from NRC come over here and let's get these populations 8
approved," because if you go ahead with this thing and then 9
you get to the end and we have not approved those populations 10 all we do is change one up, I mean, your effort can be 11 horrendous, and I don't know why we're discussing those 12 things right now, and why those things you cannot in some 13 kind of way--I mean you insist on somebody to get together 14 with you and let's settle these things up once and for all.
15 All your work becomes invalidated; your 16 inspection becomes invalidated, because you're talking about 17 different populations, different membership.
18 You've got to put new scaffolding, new 19 people got to go there.
So it's so important to get this 20 thing settled.
You got a lot of work to be done to get 21 to those items that you have randomly selected.
22 Later on, when we get together, we can 23 make some recommendations what needs to be done.
MR. HANSEL:
Again, I think if we can sit--
25 you can just write so much in the written word, you have to
135 1
talk to the engineer and see the methodology, I think we 2
can defend each.and every population and how we got there 3
through the work processes.
4 I prefer that we do that; I'm just as anxious 5
as you.
6 MR. CALVO:
Okay.
7 Does<anybody have any questions with the work 8
process?
If not, we can go to the next item.
9 MR. SHAO:
Maybe--what I suggest in the future 10 we have a meeting on civil structure, a meeting on mechanical, 11 meeting on electrical, all lumped in--
12 MR. HANSEL:
Fine.
13 MR. cALVO:
We better have a lot of meetings, 14 Larry.
15 MR. NOONAN:
Let me talk to that a little bit.
16 AfterNie caucus, I'll put on the table the 17 kinds of meetings we need to have, and then we can organize 18 when we will have them.
19 I woul.d guess maybe it would be more productive 20 to do it at the site.
I don't know.
Let me talk about it 21 later.
22 MR. CALVO:
Well, if we're finished with the 23 work process, I guess the next item on the aggenda is the 24 sampling, and Charlie has high priority; he has the question 25 first, so I think we owe him an answer.
-, ~.. - -
136 1
MR. SHAO: _First, I'd l'ike to get a large bore 2
piping--
3
... laughter.-..
4 MR. HANSEL:
Ron Tate is going to talk about 5
large bore piping configuration.
6 MR. TATE:
My name is Ron Tate; I'm discipline 7
engineer for the large bore piping configuration in the I
mechanical group.
9 There are approximately nine different popu-10 lations involved under the large bore piping configuration 11 itself.
We lumped all the attributes under the one work 12 process of installation.
13 There are numerous individual little processes 14 which may go into the completion of the pipin'g system.
In 15 the configuration, we're looking at the overall installation.
16 Under that installation, I believe there are 17 nine different attributes up there; there should be, and 18 grouped under each of those attributes are the sub-attributes 19 (I call them sub-attributes).
20 The first one is piping orientation; more 21 general, it covers the sequencing of the various components, 22 the location of the piping which would also include the 23 location of most of the components that get involved.
24 We look at the branch connections that are 25 used, the clearances that are soecified in the design
.e
137 1
. documents.
2 Beyond that, it becomes a little more specific 3
in that it talks about the various components that go into 4
making up the installation or the piping system.
It's 5
broke out to cover a few items that are specific.
6 In most of the cases we gover identification, 7
as well as the location.
8 The piping systems that were involved had I
typical attributes which I mentioned, the component sequencing 10 the location,.thef. dimensions,~ dimensions being primary, 11 linear, and the identification of components involved was 12 put together using the rules established by the ASME Code, 13 Sectim1III, with the installation being the responsibility 14 of Brown & Root and done by Brown & Root procedures, as 15 well as the inspection being the responsibility of Brown &
Root and, again, done by Brown & Root procedures, all of 17 which were developed based on the QA programs in accordance 18 with ASME III.
19 MR. SHAO:
Are you talking about just pure 20 piping, or are you including pipe supports?
21 MR. TATE:
This population only covers the 22 piping itself, pipe supports are not included.
23 MR. HANSEL:
That's a separate population.
24 MR. SAFFELL:
Large bore piping in this case 25 is the population; is that correct?
138 1
MR. TATE:
Yes, sir.
2' MR. HANSEL:
Large bore piping configuration.
3 MR. SAFFELL:
My question is you then 4
indicate a number of attributes and sub-attributes on the 5
slide, the one that you just took down.
6 My question is, what do you consider to be 7-a deficiency or defect in the population?
Is it down at the 8
sublevel like if you happen to notice the pipe size is wrong?
I Is that one defect?
10 MR. HANSEL:
- Yes, MR. SAFFELL:
So any of these items shown 12 on this particular chart--
13 MR. HANSEL:
You have to get to the sub-14 attributes to get to the defects.
15 MR. SAFFELL:
That's my question.
Any defect 16_
at the sub-attribute--
17 MR. HANSEL:
You could have an orifice or 18 flow assembly process mislocated.
That would be a sub-attribute level, that's where the defect would be written, 20 the deviation--
21 MR. SAFFELL:
And the population consists of 22 some number of piping configurations by piping analysis; 23 Ron, is that the beginning and end of a configuration?
24 MR. TATE:
The selected sample will be an 25 isometric that was produced--
L.
m 4J
-u-.
139 1
MR. SAFFELL:
I guess that isometric corres-2 ponds to the--what I call the stress iso, or is the isometric 3
characteristic of a subsystem?
4 MR. HANSEL:
I don't think we know the answer 5
to that.
6 Again, you pull isometric drawings, the 7
relationship between that and the analytical work, I don't 8
know, it may have been the same but we don't know that.
9 We look at construction and we pull from 10 all the isos.
11 MR. HOOKWAY:
The isometrics are not stress 12 marks,'are not' systems.
13 MR. TATE:
Samples are made up--or the 14 populations are made up from isometrics, which includes 15 all bore piping systems.
16-.
MR. NEVSHEMAL:
As we went through before 17 with respect to cables, you indicated that the reason cables 18 was a population was because they had three work processes that were distinct from other electrical aspects which were 20 prepull, pulling and termination.
What are the work 21 processes that make large bore piping configuration a 22 distinct population as opposed to the other eight populations 23 that you have here?
24 MR. TATE:
In the population for piping 25 configuration, we're interested in determining whether or
'l40 1
not the installer installs the system as designed, whether 2
or not he met the specification and the procedure requirements 3
for clearances and locations.
4 And looking at that, you can only look at the 5
overall installation.
Since I'm interested in where he put 6
the various components I can look at one valve, 10 strainers, 7
10 feeders, one valve; he can locate where he's supposed to.
I It's not important, the component he was looking at, or I
the attribute.
10 I '7 looking at clearances, the same would.
11 apply.
If he can prove by installation of various components 12 that he always chose the right piece of equipment, that 13 the. isometric drawing called out the right piece of equipment, 14 and I'm satisfied that he can follow the instructions good 15 for the installation.
16 MR. HOOKWAY:
Ron, excuse me.
Can you go 17 back one step to the large bore piping population; there 18 are a number of different isometrics in the plant, and some 19 are BRPs and BRHLs.
Which set of isometrics are you working 20 on?
21 MR. TATE:
The isometrics that were used in 22 piping population are the BRPs, Brown & Root piping.
23 MR. HOOKWAY:
Do you know how many total 24 BRPs there are in the plant?
25 MR. TATE:
In the population listing there
141 I
were 3,000.
2 MR..HOOKWAY:
The population you're sampling, 3
how many isos?
l 4
MR. HANSEL:
Sixty.
But, collectively, 5
you consider the--we're looking at a total of 99, when you 0
lump both samples together for large bore piping configu-7 ration.
I MR. NEVSHEMAL:
What do you mean both samples?
MR. HANSEL:
The random sample and the 10 engineer's sample.
The first sample of 60 is random.
The 11 second sample consists of those items, also selected at 12 random, that are required for safe shutdown; they can be I
60, anywhere between there and a hundred and twenty, due I4 to the overlap.
In this case we ended up with 99.
15 MR. SHAO:
You have large bore piping 16 configuration,.th'e' bolt.ed joints and the welds, are in 17 a different population?
18 MR. HANSEL:
Yes.
19 MR. SHAO:
Why?
20 MR. HANSEL:
Again, for ease of pulling those 21 for sampling.
And work processes are different is the 22 primary reason.
23 Welding is a work process as compared to 24 installation, the kinds of things we looked at here.
25 Same way with bolted joints, different work m.
142 1
process.
We could have put.it all together but it would 2
have been very complicated.
This gives us a nice, simple, 3
basis; easy to draw for. the sample, and to evaluate the 4
work process, which is the real driving force of it all.
5 MR. SHAO:
You have separate population 6
for joints and welds?
7 MR. HANSEL:
Yes.
8 I think this gets back to the questioning--
9 line of questioning earlier.
We saw enough differences 10 in the work processes when we went to some other populations.
11 MR. HOOKWAY:
One of your subpopulations is 12 Moment Restraints.
Can you define that, please?
13 MR. TATE:
Moment Restraints for forgings, s
14 component part of the pressure boundary of the piping system, 15 they will have, generally, the designs I've seen, either a 16 round forging with a machined off section on each end, or 17 maybe a square forging is hollowed out.
II The moment restraint portions that I checked 19 is only the pressure retaining part; the second half of it 20 is bolted back to the structure to take the first moment 21 out of the said piece of equipment, bolted up to the restraint.
22 MR. NEVSHEMAL; Let me ask one more thing.
23 With respect to large bore piping, are you going to look at 24 60 samples, or 60 randomly selected situations?
Is that
- 25 the level at which you're going to create the confidence of
}
143 1
95-5, or what level--
2 MR.. HANSEL:
On large bore piping configuration 3
we pull a sample of 60, randomly.
That's the basis for the 4
95-95.
5 We then continue to select randomly and test 6
to see if the items selected is required for safe shutdown, 7
until we assure ourselves that we have 60 of those, either I
in the first population or the second draw.
9 In this case, for this configuration, large 10 bore piping configuration, 99 items that we will look at.
11 MR. NEVSHEMAL:
So you can say the same 12 thing with respect to HVAC ducts?
I
(
MR. HANSEL:
Yes.
MR. NEVSHEMAL:
You're going to create the 15 95-5 at that level?
' 0 MR. HANSEL:
Yes.
MR. SHAO:
Where does piping attachment fitting a 18 go?
MR. HANSEL:
Al, you want to address that?
20 MR. PATTERSON:
Pipe attachments go with the 21 support itself, excluded from the piping population.
22 MR. SHAO:
The piping attachment is not in 23 the piping configuration?
MR. PATTERSON :
One of the three piping 25 support populations.
144 1
MR.. HANSEL:
Which you now see here on this 2
chart.
3 MR. TAREO Did I also see in civil structural 4
embedments--was that covered?
5 MR. HANSEL:
Yes.
6 It's in with the concrete population, and 7
I guess it may be also in the structural steel.
8 Are embedments included in structural steel?
9 MR. PATTERSON:
Attachment plates for supports, 10 yes.
11 However; they are attached to the support 12 itself as an integral part of the drawing.
MS. EARLY:
We are' going t6-take a short
'14 recess but, first, does anyone have any further comments?
15 MR. CALVO:
Yes.
The next subject is going 16 to be sampling.
Let's get a feeling how long it's going 17 to take.
18 Do you have a presentation prepared?
19 MR. HANSEL:
No.
MR. CALVO:
So it will just be questions 21 and answers?
22 MR. HANSEL:
Yes.
23 MS. EARLY; Okay, we will have a short 24 10-minute break.
25 IShort recess.]
145
.. ~.
1-MR..NOONAN; Just for some planning purposes, 2
we can continue on with.this part of the discussion.
I 3
would guess we could stop right around noontime.
4 I'm going to caucus with the staff at 1:30, 5
and I think we can resume our meeting.
I'll need at least 6
half an hour, maybe 40 minutes.
Maybe around 2:15 we 7
can reschedule the meeting, start back around about 2:15.
8 MR. COUNSIL:
I would just suggest we don't 9
need an hour and a half for lunch.
I don't even need an 10 hour1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />.
There's nothing around here you can eat in an hour.
11
... laughter...
12 MR. NOONAN:
I need an hour and a half.
13 We'll caucus with the staff at 1:30 and start 14 the meeting back around 2:15.
15 MS. EARLY:
I'd like to turn the meeting 16.
over to Jose.
17 M 1.
CALVO:
Let me start and maybe stimulate 18 discussion.
19 I guess we want to know why 95-5; why not 20 95-1 or 96-2 or 99-1.
Also, we'd like to know the random 21 sample based on population; we missed the population by one 22 or by two.
Would that invalidate your sampling?
You have 23 to start all over again? because you come out with different 24 numbers?
25 so how about telling us about it.
1AC 1
Also we owe an answer to Charlie; he's not 2
here.
3 MR. HANSEL:
Okay.
4 Let me lay the groundwork with some preliminary 5
discussion then I think we can best get at it through 6
questions and answers.
7 There's nothing magic about the 95-95.
It 8
is a fairly recognized quote standard, standard in the 9
industry; it's been used before.
I have used it before 10 and am very comfortable with it.
11 Again, with my knowledge of inspection and 12 practices that go into the building of the plant, it's
~
13 probably adequate in terms of evaluating construction 14 adequacy, which is really what we're after.
15 Yes, you could have gone for a much larger I0 sample, but I don't know that it would have given us that 17 much more of a comfort' zone in terms of the adequacy of the II plant.
II So the selection of 95-95 came from a recom-mendation of myself, discussed at length between myself 21 and the CPRT, senior review team and TUGCO, and that's about 22 all the justification I can give you for it.
23 It's an accepted one that's.been used and 24 proven and I feel that it, gies you a good evaluation plan.
25 The thing that you have to keep in mind is
-a a
w
147 1
that when we make a judgment about the adequacy construction, the inspections and the documentation reviews are one source 3
of input.
We have done, and are doing, a detail review of 4
the construction processes, the procedures for craft and for 5
inspection.
6 We understand what's going on in the world of 7
training; we know what took place in the area of process 8
controls.
9 We know what took place in terms of work 10 control during the actual installation fabrication processes.
11 We have to take all that into consideration.
12 Nobody can inspect the quality into the hardware.
I think 13 the country has learned that now over the last couple of 14 years.
15 You cannot inspect quality into the hardware; you have to build it in there and you have to control it 17 through processes, and througn people.
18 Inspection is just one more check or verifi-19 cation as to the adequacy of that.
20 So if you take all that into context, I 21 feel comfortable with the 95-95.
22 Provided we have good homogeneous populations and we have good sampling techniques.
And I think that we 2d have that.
We've been very, very careful, very careful to
o 148 I
the point that it's actually cost us time to be sure that 2
we have everything in the population that should be in the 3
population.
4 That takes effort.
We have had to take and 5
extrap from drawings, manually, various items of equipment, 6
but we have made certain that the population was all-7 inclusive.
f I
We have implemented and monitored the sampling I
process with extreme care, to be sure that there were no 10 biases, and that we did have a truly random sample.
11 And, again, back to one of the points you 12 brought up, we concentrated on work processes, and if I 13 verify the work process, then I should not have to be too 14 concerned about did I get some of every kind of equipment, 15 because I'm verifying the process.
That can be argued from 4
now on.
17 But we feel comfortable that'we've got good 38 homogeneous populations, and that we control the sample 19 selection process from there forward.
20 Again, the number that I gave you earlier, 21 we're looking at a large sample of attributes in that plant, through sampling work processes, about a half a 22 23 million, somewhere in that neighborhood.
I suspect it may 24 even go bigger than that.
That's a first cut.
25 So this program, this plant that we're
149 1
looking at is undergoing, in my opinion, probably the most 2
extensive review of any in the country, in terms of what 3
we're doing in the approach.
4 I think we've got a good program.
5 That's about all I can give you on the 95-95 in 6
terms of how we got there and--
7 MR. SHAO:
When you say other industry, what 8
other industry?.
I MR. HANSEL:
Other plants in the industry.
10 Braidwood is based on 95-95.
II MR. SHAO:
Diablo Canyon?
12 MR. HANSEL:
Diablo Canyon, they use 95-95.
13 MR. NEVSHEMAL:
John, I'm not a statistician 14 so when I hear 95-5 or 95-95, what specifically are you 15 saying in those two numbers?
16 MR. HANSEL:
We have a 95 percent competence 17 level, that there are no more than five percent deviations 18 in the plant undetected.
There are numbers somewhat higher than that and I don't knou how much higher because our 20 sample is based on 60, when you take both samples combined.
21 But it's even bigger-than that combined number when you take--
22 On some attributes, I'll be looking at welding, for 23 instance, I'll be looking at a number of populations.
24 I'll be looking at hilty Isic] bolts in a 25 number of populations.
I'll be looking at drilled holes in e
,ww--
y
150 1
a number--
I'll probably be testing, looking at 3,000 hilty 2
bolts.
That's an estimate.
3 I hate to estimate the number of welds we'll 4
look at.
I would estimate we'll look at five or six thousand 5
6 So you can't just zero in on the sample of 7
60 that says you're going to look at 60 items; you're going 8
to go to that item, and that item may have six to ten welds.
I MR. NEVSHEMAL:
So what you're saying, then, 10 within certain attributes what processes you may have higher 11 confidence--
I I
MR. HANSEL:
Yes, but I don't know how to 13 calculate that without going back and doing a lot of, you 14 know, numbering.
I Again, I wouldn't want to.
I'm concentrating 16' on the work process and I'm gathering data to evaluate 17 that work process, gathering data.
18 MR. HALE:
John, on y'our original sample 19
.you're taking 60, and on the engineered sample, you're 20 taking another 60 which may be common to the first.
Those 21 two samples are looked at in the same way, using the same 22 inspection criteria?
23 MR. HANSEL:
Yes.
24 MR.. HALE:
If a defect is identified in 25 either one ~of those ~ samples and you expand it by 35, you
e 1
~
s k[( /
~l' 151 1
look at the same criteria there,.use the same attributes 2
in that re-inspection of that additional 35?
q 3
MR. HANSEL-The first 35, yes. Beyond that, 4
,'d'ependin,g on how many discrepancies we found and what the
\\
l 5'
root cause and generic implications tell me, and if I have
'J-
?
6-repe'ats--if you want to get into that, I think with 7
Charlie's question we can--then I may go off and look for I
just that attribute, but that's a whole chain--
In fact, 9
now may be a good time to discuss that.
10 MR. CALVO:
It's not as pending--not based
'II L
.tn1 deviation, so the safety significance of the deviation; 12 correct?
MR. HANSEL:
Yes.
14 MR. CALVO:
So you may find a lot'of devia-15 tions but you still cannot do more than 60 if you determine 16 to make a judgment there's no safety significance.
17 MR. HANSEL:
Exactly.
Glad you made the point.
18 Because we only expand if they are safety significant.
- ' 19 i
r MR. SAFFELL:
Then you pursue only that 20 attribu,te in the' expansion; is that correct?
21 MR. HANSEL:
Not true.
The first sample, 22 we would look for all attributes, same checklist.
23 Let's go through that process.
24 I'm going to ask Al Patterson to do that (25 i
i
, since he's far more familiar than I am.
M J
\\
152 1
MR. PATTERSON:
The first sample we expend 2
a total of 35, let's assume the total population is over a 3
hundred.
4 And at that time we assess the same, use 5
the same checklist, the same procedures, same quality 6
instructions that we've written for each one of our popula-7 tions.
8 When we ha.ve a criteria that's violated, 1
9 based on the original checklist and the original research 10 with the architect / engineer requirement, and we get a safety f
l 11 significant deficiency again, the second time, we determine i
12 if that was the same root cause as was the first one.
13 If it was, we use only the attributes as we've l
stated in a procedure that we'Vr written to ourselves, only
(
15 the attributes necessary fg 41.;;ify deficiencies of that 16 type, verified in subsequent re-inspection.
17 If we have a different type of safety 18 significant deviation, a different reason, we will use the same checklist again in its entirety.
I 20 MR. TRAMMELL:
Again for what?
21 MR. HANSEL:
For everything; all attributes.
22 If you have differences in the safety signifi-23 cance, the root cause is different, then you'd go for full 24 inspection of all attributes.
25 MR. SAFFELL:
Are you talking about a hundred
153 o.
1 I
percent at that point?.
2 MR. PATTERSON:
No;.we're still talking about 3
the additional 35 that I said was the first expansion.
4 MR. TRAMMELL:
You did your additional 35 5
on-another hit; now you have another hit--
0 MR. PATTERSON:
After the second expansion 7
you go to a hundred percent of all the samples in the popu-8 lation.
I MR. HALE:
After three safety significant 10 defects you go to a hundred percent?
II MR. HANSEL:
Yes.
First sample, plus two 12 expansions.
13 MR. SAFFELL:
But I did hear you say that 14 when you have to expand the sample, whether it be the first 15 or the second--but when you have to expand the sample, you 16 look at all attributes, not just the attribute that caused 17 the deviation?
18-MR. HANSEL:
Let me clarify that.
The first expansion you go for all attributes.
Now if you don't 20 find anything more you're in good shape.
If you find two, 21 you're going to expand the second time then.
~
22 If the root cause was the same you would go 23 for that attribute; if it was not the same you'd go full 24 checklist.
25 MR. SAFFELL:
I understand.
154 1
MR. TRAMMELL:
If you find nothing in that 2
second, that 31, then you accept--
3 MR. HANSEL:
Accept that sample.
4 MR. TRAMMELL:
And if you find one or more 5
in that last group of 31, you decide to do a hundred percent.
6 Both units.
7 MR. HANSEL:
If the sample is all-inclusive 8
of, you know, one, two and common.
9 MR. CALVO:
You have 23 populations; you 10 found nothing wrong, but if you found something wrong in 11 one population, the way you have combined them, you're going 12 to have a hell of a big--
Excuse me.
13
... laughter...
14 You're going to have a big effort to go through 15 all those items in the population, so you've got to trade 16 off to have small populations or something.
17 MR. HANSEL:
Yes.
18 MR. NEVSHEMAL:
.Does the expansion that you're 19 speaking of with respect to the number that you're adding 20 up, is that to maintain the 95-5?
21 MR. HANSEL:
Yes.
22 MR. NEVSHEMAL; Because, as I understand it, 23 what we're saying here is that there's a certain number of 24 samples of which you must be deviation free before you've--
25 in order to maintain the 95-5?
p.
y
~
..-,y
155 I
MR. HANSEL:
Y e s.-
It's built around maintaining 2
the 95.5.
3 MR. NEVSHEMAL:
So that means that you have 4
to maintain, you have to look at all the attributes and 5
all these expansions until yon get to some point where you 6
say we're going to have to do a hundred percent.
7 MR. HANSEL:
Or after the second sample, if 8
I have two failures of the same attribute and nothing dif-9 ferent, they're both safety significant, we're going to go 10 after that attribute.
11~
MR. NEVSHEMAL:
The reason I'm asking this 12 is because you can increase samples and have more allowable 13 defects; you continue to increase until you have a certain 14 number of allowable defects and then say I still have 95-5.
15 I think I heard you right, yes, MR. HANSEL:
16 I believe so.
17 MR. TRAMMELL:
Are you going to correct the 18 defects that you find?
19-MR. HANSEL:
Everything that I've identified 20 in deviation reports are given to the project and NCRs are 21 written.
22 MR. TRAMMELL:
On defects or deviations?
23 MR. HANSEL:
Deviations, yes.
24 MR. TRAMMELL:
Whether safety significant or 25 not--
156 1
MR.. HANSEL:
All of them.
2 MR. TRAMMELL:
--they will be corrected?
.3 MR.' HANSEL:
Yes.
They will review those, 4
and repairs are made--
5 MR. TRAMMELL:
NCRs referred to the review 6
process may or may not be corrected, depending on how its 7
disposition?
I MR. HANSEL:
Yes.
I But every deviation that we write that after 4
10 it has passed the check to determine it's valid, we couldn't 11 find any other piece of engineering paper work that said 12 that that condition was acceptable, go to the project and 13 there's a one-for-one NCR for each deviation report.
I4 MR. HALE:
Mr. Hansel, I seem to recall reading 15 in the action plan that you didn't go to a hundred percent I0 re-inspection of population until 10 safety significant 17 defects were identified.
MR. HANSEL:
I don't know where you read that.
It's not true.
20 MR. HALE:
I didn't read it.
21 MR. HANSEL:
You should know me better than 22 that, Cliff.
23 MR. CALVO:
The same subject, I've got a 24 question about this exploratory sampling that you performed 25 by the CPRT, what effect that has on the sampling process.
l 147 1
You know, prior to starting your work up 2
there you did do some exploratory.
So I checked this over.
3 Now how do you factor that?
Can you talk 4
about it?
5 MR. HANSEL:
I'm not sure I'm understanding 0
exploratory work before we started our sampling.
7 If you're talking about the review of the 8
drawings and codes and standards and the work processes--
MR. CALVO:
Yeah.
Larry, this question is 10 one of your questions on the plan, the, exploratory sampling 11 performed by the CPRT on the overall sampling process.
12 Can you--
13 MR. SHAO:
[ Inaudible ~. ]
14 MR. HANSEL:
Maybe I can clear it up; you 15 may be ta1 king about the accessibility walkdown?
16 MR. CALVO:
No.
17 MR. HOOKWAY:
Refresh my memory, Jose.
18 MR. SAFFELL:
There are words in the program 19 plan that talk about exploratory sampling, maybe not 20 exploratory sampling, but--
21 MR. CALVO:
That word is used.
22 MR. SAFFELL:
I think it is.
Exploratory 23 sampling.
24 MR. MALONSON:
I'm working off the top of my 25 head now but I remember the comment it pertained to a couple
158
'l-of action plans'that described exploratory work over in the 2
background section, tha't actually should have been described L3 in the methodology.,
c2.
4 We felt that the' work done described as 5
exploratory was part of your physical CPRT effort in' 0
methodology that satisfied the plan, I don't remember which 7-ones'they were.
I-MR. HANSEL:
I'm not' totally with you but I
let me start out and see if I can help.
10 In the TRT issues, and those issues identified 11
~by external sources, we'have individual action plans on 12 those.
13 And there are inspections and' documentation 14 reviews being. conducted to resolve those issues, to-investi-15 sate to resolve those issues.
16 The engineers associated with our self-17 initiated program are aware of the results of that and what 18
.is happening.
I
Now in terms of exploratory for self-initiated 20 program, our engineers walked the plant and became familiar, 21 but there were no inspections conducted, other than those 22 associated with our samples.
23' There's one other point that may help you.
24' For each' inspection checklist quality instruction that is
- 25 written by the engineers, we have a requirement that the l
l 159 r_
1 engineer and the inspectors walk that process down, there's 2
classroom training, indoctrination, and there is a walkdown 3
on the hardware with.the engineers to talk through that 4
procedure.
5 There have been changes to the procedure-6 based on those walkdowns and basically OJT.
Is the 7
inspector happy with the procedure; can he work with it, 8
does he understand it.
Is it clear.
9 And that's a mandatory requirement that that 10 be done on every population.
11 There have been changes to those inspection 12 procedures as a result of that, wh'en you actually get to 13 the hardware.
14 MR. CALVO:
Everybody's happy?
15
[No response.]
16 MR. CALVO:
Okay.
I guess the other subject 17 that we're interested in is human reliability.
18 Burney, can you talk about that a little bit?
19 MR. SAFFELL:
Well, the comment stems from 20 the fact that you have people that are, albeit trained to 21 same level, are still--can conceivably make mistakes.
22 MR. HANSEL:
They're people; they're human.
23 MR. SAFFELL:
That's right; they're human 24 beings.
25 And the question comes from--is that--you
]
160
-. ~. - -.
I could adjust the samples, I believe.
2 Again, I'm not a statistician, but I believe 3
you could adjust your sampling to accommodate a certain 4
assumed level of human error, if you will, liability.
5 The question is, I don't think that was 6
done; what is your rationale?
~
7 MR. HANSEL:
Again, I understand exactly 8
where you're at.
I've done some of those studies on 9
inspector repeatability.
I've had some done, and I under-10 stand the situation well.
11 We chose to not do that, didn't feel it was 12 necessary.
I3 Again, a sample of 60 on a 95-95 basis provides I4 us with one set of input about that plant, and it's a good 15 measure of the adequacy of the construction and that work II process that we're sampling.
17 That, plus my knowledge, or our knowledge 18 of the specifications and the process controls and the other 19 procedures that were followed will be our total input for.
20 drawing conclusions on each population, each work process.
21 Plus, the fact that you just cannot inspect 22 it into it--if I took--I don't know what the number is, but 23 that added samples might only add another five or ten.
I 24 don't know what the number is.
That doesn't do a whole lot 25 to warm up my juices.
I have a 95-95 program and I'm
161 1
perfectly comfortable with that.
2
-And I don't.think that the human reliability 3
factor or the inaccuracies of inspection, if you want to 4
call it, do a whole lot to that.
I don't think there's 5
that much of an effect.
6 We're not doing inspections of record; the 7
hardware's been inspected many times.
O I could run out there next week with another 9
group of inspectors and probably find some different things.
10 And the next week find some different things.
11 And that doesn't help. I've got good well-12 trained inspectors; they're independent, evaluating the 13 work process, at an adequate level.
I4 MR. CALVO:
The other question that we asked 15 about, the random sampling, based on the total population.
I6 What is the sensitivity there?
If you talk to numbers, how I
much of that random--
18 If I've got a population based on X number of 19-items or occupants in a population, I think the way you 20 described how you randomly sample, cascade the numbers, 21 based on previous numbers, divided by the total, something 22 like that, what will happen--say you miss it by five, or 23 you miss it by ten occupants, how does that change your 24 random sample?
You have to start all over again?
25 MR. HANSEL:
I'm not reading you, Jose.
162 1
Are you asking would it change the sample size?
2 MR. CALVO:
Not.the sample size; the. location 3
of those items.
4 MR. HANSEL:
Are you talking about accessibil-5 ity?
6 MR. CALVo:
No.
If I got a population, for 7
instance, like cable, 14,000 cables, and then you say based 8
on 14,060 items, I must go to the 41 switch gear, I'm 9
going to go there by the control room, depending on the 10 items that you have selected by the random sample, based 11 on 14,000.
Then one of your. engineers comes to you and 12 s'ays it's not 14,000; it's actually 14,010; how would 13 that affect the random sampling?
Are you going to go end 14 up in different areas of the plant?
You got to start all 15 over again?
16 MR. HANSEL:
No.
17 MR. CALVO:
What is the sensitivity?
II MR. HANSEL:
Well, again, you take the total 19 items in the population and they are sequential in numbers; E
anci then we crank tne random number generator and we select 21 from that.
22 So if there were 10 more or 20 more or 30 23 more, it probably wouldn't make a bit of difference.
24 MR, CALVO:
I. bel.ieve your calculation is based on the total?
And if you change the total you change
,7
163 1
1 1
the--
See, if I got one item right here by the lake and 2
one up in the elevation--
3 MR. HANSEL:
let me ask Fred Webster, our 4
statisticials, - to address that.
5 MR. WEBSTER:
Fred Webster.
You're right 6
that the population is very integral in selecting the exact 7
random sample that you came up with.
8 Sometimes we do find that items are left out 9
of a population description and they're found later on.
10 What we've been doing with that in general 11 is to select a second random sample from the smaller sub-12 population, if you vill, on a proportionate basis, so that 13 if you found, say, 10 percent more items and didn't have 14 them in the first population, you'd sample an extra six.
15 The idea is that, yeah, you can change the 16 number by one and come up with a completely different random 17 sample; but the idea of the random sample is for each and 18 every item to have an equal chance of being selected.
19 MR. TRAMMELL:
It looks like that these 20 numbers come close to the numbers I recall from the distri-21 bution for an infinite population--
22 MR. WEBSTER:
You're exactly right.
13 MR. HANSEL:
You're right on.
24 MR. WEBSTER:
And because of that it is 25 conservative.
v 0
164 1
MR. TRAMMELL:
So is this based on an infinite 2
population, or is.it actually sensitive to--
3 MR. WEBSTER:
The size, the number of 60 is f
4 based on an infinite population.
5 The exact item that's selected is based on 0
the population definition, a thousand items, or ten thousand 7
items, that number is an integral part of selecting the I
exact rcndom sample that you look at.
I MR. TRAMMELL:
And if your random sample--just 10 for the record, if your random sample should pick the same i
11 item twice, do you reject it and go on to something else?
MR. WEBSTER:
Yes.
13 MR. NEVSHEMAL:
In the mechanical area, with 14 respect to mechanical components, how are you going to 15 create your list of random items to look at?
For example, 16 are you going to throw in individual P& Ids (phonetic) and 17 then select them randomly and look for bumps or whatever 18 that are going to be on there, or are you going to make 19 an individual pump of random items?
20 MR. HANSEL:
If I understand the question, 21 the mechanical equipment items list in the population will 22 include all mechanical items.
23 MR. NEVSHEMAL:
So they're individual; 24 they're not--like, for example, if you might have four 25 pump--
165 I
1-MR.. HANSEL:
It's all, because the work 2
process for installing mechanical. items is the same.
It's 3
all.
4 MR. HALE:
How does the inaccessibility of 5
an item affect the random nature of your selection?
6 MR. HANSEL:
We are very sensitive to that.
7 We've not had a problem yet.
8 Let me address that in two points.
One is, 9
there's no reason to believe that the control of the work 10 process and the inspection process is any different on an 11 item that was inaccessible versus an item that was accessible.
12 As we draw our samples and do our accessibility 13 walkdowns we're in continual discussion with Fred Webster, 14 the statistician, to determine if, in fact, we're ascewing 15 that in any way, aimed at more unit one versus unit two, or 16 are te bias in any way.
And so far we've not had a problem 17 But we're very sensitive to that.
We watch 18 it on a population-by-population basis.
19 MR. HALE:
Is that why you've not had any 20 accessibility problems?
21 MR. HANSEL:
We've had accessibility problems, 22 but nothing that would give us ascewed distribution.
23 TRAMMELL:
Have you rejected any inaccessible 24 samples you selected because of that?
25 MR. HANSEL:
Oh, yes.
Yes, we draw our first
.,,,v--
n-
166 I
sample and we go do a walkdown; is that hardware accessible, 2
is it acceptable to all attributes.
If it is available, 3
if it's accessible, it stays in the population.
4 If not, we throw it out and draw another 5
random.
6 MR. HALE:
Question, all attributes or 80 7
percent of the attributes, or 60 percent of the attributes?
8 MR. HANSEL:
The target is to get all of 9
the attributes.
10 MR. SHAO:
How do you check rebars?
11 MR. HANSEL:
Rebar, we declare right up front 12 as inaccessible.
13 MR. SHAO:
Can you check the copy placement, l
14 copy placement package to see whether rebar was in place?
-15 MR. HANSEL:
Through documentation reviews, 16 yes.
17 MR. SHAO:
Did your people take photographs 18 during the computation?
19 MR. HANSEL:
I don't think photographs were 20 taken, but th'ere's documentation in there of an inspection 21 verification of the placement of rebar, spacing, timing, 22 l
location, size.
23 MR. HALE:
John, you didn't answer my question.
24 MR. CALVO:
That's what I was, going to ask; 25 can you repeat the question?
\\
L_
167 1
MR. HALE:
You said you target all attributes, 2
but you didn't say all attributes; and I'm wondering what 3
your criteria for rejecting something that's inaccessible is.
4 MR. HANSEL:
I, guess--at this point, Cliff, 5
we don't have a hard and fast ground rule; we'd like to get 6
them all.
7 Patterson has been instructed with his guys 8
to be nervous if we get below the 80 percent, and we want 9
to talk about it.
We probably should go after another sample.
10 That's not a hard fast ground rule.
It's 11 going to depend on the population and what you're looking at.
12 We're sensitive to that, we monitor it, 13 anything that gets us nervous, you know, low, we'll add 14 another sample.
15 The real intent is when we finish that we 16' will have looked at at least 60 items for each work process, 17 and we're aiming to get as many of those attributes for that 18 work process as possible.
19 MR. HALE:
Thank you.
20 MR. HANSEL:
Again, I have to draw conclusions
-21 on the hardware if I can't get enough attributes to measure 22 it, then I'm going to go after some more, and I can't say 13 where that line of sand is.
24 MR. CALVO When do you make that determination?
15 before your inspection, right?
168 I
MR. HANSEL:
Normally, yes.
But you may 2
have a case where something has occurred, particularly in 3
Unit II where construction is going on, I may have access 4
today to a piece of hardware, by the time I get to it next 5
week I may not have.
6 We've tried to work with the project to avoid 7
that, and it's happened in very few cases.
I But it's before the inspection in most cases.
9 We've had--
Well, that's enough.
10 MR. NEVSHEMAL:
If I understand what you're 11 sayins, when fou've selected--and let's use the large bore 12 piping configuration because we all have that sheet--if I I3 was to select a iso and that iso ended up not having eccentric I4 reducers, strainers on it, valves, you might consider saying 15 I need a different iso because I don't have 80 percent of 16 these attributes represented on that iso.
Is that what 17 you're saying?
18 MR. HANSEL:
That's the intent, but you have 19 to realize that several of those attributes in that population,
20 or very few of them, are flat.
21 Again, I'm looking at the work process, and 22 the process of installing much of those kinds of equipment 23 listed across the attribute line are the same, so I don't 24 care if I have all moment restraints, or all strainers, or 25 all screw joints.
I'm looking at the work process.
169 1
I don't know if I answered your question or 2
not.
3 MR. NEVSHEMAL:
You answered it in great 4
confusion in my mind; sorry.
5 MR. CALVO:
Any other questions on sampling?
6 MR. SAFFELL:
I do have one other question 7
on it.
8 In the case of something like the large bore 9
piping where you do have-(this is a followup of John's 10 question), where you do have a large number of attributes 11 it seems to me that it's very, very possible that you will 12 draw your sample and go through the entire process and 13 actually not address, or not have any of your isos had one, 14 two, three'of those attributes there, did you check that 15 kind of thing?
16 MR. HANSEL:
No.
Because, again, I'm after 17 work process.
II Let me just talk about that particular popu-19 lation.
Ron, you can correct me if I go astray.
20 You have basic' ally seven kinds of hardware 21 there, plus orientation.
By the time you pull 60 isos and 22 the average run length of an iso, Ron, in that population 23 is how long?
24 MR. TATE:
You can't stick an average on an 25 isometric because it may range from--an isometric may be a
170 1
relief valve, and from there'it may,go to 400 feet of pipe.
2 MR. HANSEL:
So you're going to get a large 3
number of ccmponents to look at installation in that popula-4 tion.
I don't know what the number is, but it's probably 5
a couple of hundred, where you can look at component instal-6 1ations of the kinds listed there and for those sub-attributes 7
primarily location, orientation, size, elevation, et cetera.
8 Again, you're worried about the work process.
I And the work process for installing those 10 seven kinds of hardware are the same.
11 MR. HOOKWAY:
Do you know off the top of your 12 head if there are any expansion joints in the population I3 l'
of the last inspection?
14 MR. TATE:
There was one expansion joint that 15 was not of a metallic type.
We chose it as a sample 16 originally.
It was rejected because it was not complete.
17 It was the only one that was chosen.
18 MR. HANSEL:
Again, we're not inspecting the 19 expansion joint per se.
We're looking at the installation 20 of that expansion joint.
21 MR. MALONSON:
There was an open issue con-22 cerning expansion joint clean 1.in.e.ss
.Wil-1.that..open, issue be 23 aligned--not an open issue, excuse me--there was an external 24 source issue pertaining to.the expansion joints.
Will that 25 condition be assessed in one or the other of your piping
- - ~
o 171 i
1 inspection?
2 MR. HANSEL:
That's being looked at in the 3
issue plan on housekeeping and cleanliness.
4 MR. MALONSON:
But should it"be considered 5
in your--
6 MR. HANSEL:
If one of those expansion joints 7
end up as an item associated with mechanical equipment, 8
we would inspect it.
9 MR. MALONSON:
That may be the same [ coughing]
10 for--is that the way it may be considered in the overall--in 11 the CPRT?
12 MR. HANSEL:
I didn't read your total question, 13 Jim.
14 MR. MALONSON:
We had a similar condition 15 for other external source issues.
Will that type of thing 16 only be considered if that item shows up in the sample that 17 you selected in that inspection?
18 Take the expansi6n. joint and call it something 19 else.
It's the something else that doesn't show up as an 20 attribute in the inspection, how will you' consider that 21 other open issue?
22 MR. HANSEL:
On the issues identified by 23 external sources, a lot of those we have separrte individual 24 action plans on that will require both inspeculon and 25 documentation review.
i 172 1
MR. RALONSON:
Excuse me.
I'm referr.ing to 2
the other SSER that isn't included.
3 MR HANSEL:
Yes.
4 A lot of those are covered in various popula-5 tions of our self-initiated program.
6 MR. MALONSON:
So it will only be considered 7
when an item in the population shows up in the sampling?
8 MR. EANSEL:
Yes.
9 MR. MALONSON:
Otherwise, they'll be assessed 10 for--
11 MR. HANSEL:.Again, I'm looking at work 12 processes, again.
I'm looking at work processes.
13 MR. COUNSIL:
John, may I correct something.
14 You're right on what your work is doing, but going back--
15 Are you listening over there?
16 MR. MALONSON:
Yes.
Excuse me.
17 MR. COUNSIL:
Thank you.
I've got an answer 18 to your question.
19 His work starts after the issues--the specific 20 action plans.
Those external issues which you're discussing, 21 expansion joint cleanliness, or whatever, would be covered 22 in an issue specific action plan.
His sampling starts after 23 that.
His is a totally random sampling.
24 Those other external issues are.being addressed 25 by CPRT, but under a different umbrella.
All'are being
+
r
~
a.,-,-c.---.,-
~ - - - - - - -, -. - - -. - - -. -
-m-
o 173 1
addressed.
2 MR. MALONSON; But they still fit into 3
construction of QAOCI.
4 MR. COUNSIL:
That is entirely true, but 5
his is a totally random sample of the work processes.
6 MR. HANSEL:
Let me help--I think Larry just 7
said it correctly.
8 We have a matrix of every external source 9
issue that we're aware of, and I think we have them all.
O In that matrix I address where it came from and where we're 11 looking at it, be it either in this self-initiated program 12 or in some other issue plan or in both.
^
13 So I do have an accounting document at site 14 that talks about all of those and where we're looking at it, 15 from our aspects, QAQC and construction adequacy.
16 So I'd have to talk aspecific and tell you 17 where it's covered.
18 MR. SAFFELL:
Am I correct in assuming that this is another example of a case where, because of an external source issue, the sample size is really being 21 supplemented or is actually greater than maybe strictly 22 the random sample?
23 MR. HANSEL:
Let me give you a good example, 24 We have p'pulations on pipe supports; we Pipe supports.
o 25 also have gone and inspected those pipe supports that were n
174 L
4' 1
identified by the technical review team.
So I have that 2
data cn1 top of what I've already done.
3 And there are other cases just like that; 4
cables, terminations, a number of them.
5 MR. CALVO:
I know what you're saying; I 6
know what you're going to do.
7 But I think you're cutting down your flexibility, I
because, depending on what the external source issue brings to your attention, bring to your attention the generic 10 implication that may cross all disciplines, and you've got 11 in place a self-initiated review who may very well be adopted to take care of that situation.
13 So, otherwise, you will be doing that--that I
external source issue may-prompt you to do another self-15 initiated review of the same you did in this one, if you 16 cross disciplines.
17 So are you saying that you're going to handle 18 that independently--well, you've made that with a lot of 19 self-initiated reviews that result in generic implications.
~
I don't know if you want to say that or not.
MR. SHAO:
I think what Jose is trying to 22 say is, I think you should be aware of all the external 23 source issues when you do the sampling also.
24 MR. HANSEL:
I am.
And so are the individual 25 lead engineers; that's been discussed with them as it affects
c o
17R 1
their population.
MS. EARLY; Okay, I think it's about tbme 3
that we give Ms. Ellis a chance to speak before we go to 4
lunch, and have our caucus, so that we can consider her 5
comments during our caucus.
MS. ELLIS:
I think generally what we have is a lot of questions; and in line with Mr. Beck's comments 8
yesterday I think that we will at this time forego raising 9
all those questions.
There's an awful lot of them anyway.
10
,It would take too long to really.go into them here.
l 11 Alse I'd like to state that we believe what 12 we have been doing in the past should have been an assistance 13 l
to the Applicants; if they haven't considered it to be, we're sorry.
15 However, the past questions that we've raised, l
16 Applicants have now been aware of for some time and we still 17 haven't received answers to those.
18 So I think that what we have done in the 19 past has been an effort to assist the Applicants, rather 20 than to cross-examine their witnesses or potential witnesses, 21 or anything of that sort.
22 But we will have additional questions later, 23 in writing, which we will file formally in the hearings, 24 and there will be quite a few of them from this morning 25 and from yesterday.
17K 1
Thank you.
2 MS EARLY:
Okay, we're ready to break for 3
lunch.
4 The NRC staff will meet back here at about 5
1:15 for caucus, and we'll all be back here at 2 o' clock.
6
[Whereupon, at 11:50 a.m.,
the hearing was 7
recessed, to be resumed at 2:00 p.m.]
8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
~
4 177 1
AFTERNOON SESSION 2
2:00 p.m.
3 MR. NOONAN:
The staff had its caucus and 4
I'd like to come and sort of give you the opinion of the 5
staff and what they've decided on.
6 If you'll bear with me, some of this I'm 7
going to read, because I had to write it down while we.were 8
working and I want to put it back just like it was said in 9
the meeting.
10 In relationship to the piping and pipe 11 support issues the NRC staff feels that the Stone & Webster
- 12 work is on track.
The staff will continue the audit process 13 with Stone & Webster and probably in the office--I understand 14 probably the best place to have these kinds of audits is 15 in the New York offices.
I When Stone & Webster issues the procedures 17 in the new few weeks, we will review them very carefully; II we also understand that TUGCO, Stone & Webster, and the 19 construction adequacy program will address all the attributes 20 for the 79-14 review.
21 Stone & Webster should also be aware of all 22 the external source issues so that the re-analysis of the 23 piping reflects the work and can be more effective.
I guess, in summary, the staff is well 25 satisfied with the Stone & Webster effort.
L
178 1
When we get to.the construction adequacy, 2
we still have some questions that we need to resolve, and 3
I guess Bill Counsil--what I want to do with the staff is 4
to, in the next few weeks, at the erd of this month of 5
october, I'm going to bring the staff onsite to do site 6
audits of John Hansel's records, and so forth.
7 I'm going to kind of go through the list 8
here, John, that they gave me so you can write it down.
9 I guess there's a general acceptance to 10 the sampling techniques for demonstration of construction 11 adequacy.
When we get to the work processes the staff feels 12 that they need to look at your records and get additional 13 information before we can come to a final conclusion.
14 The information we'll be looking at in the 15 next few weeks, and for~the record,.will be the attributes 16 and acceptance criteria, codes and standards, organization, 17 crafts, and inspection controls.
18 Jose's people will be looking at basically 19 the first population item, how it was installed in Comanche 20 Peak, to the last item and how that was installed.
21 They'll look at the mechanics of the process 22 for extracting selected items for sampling.
23 MR. HANSEL:
sorry, Vince; would you repeat 24 that one?
25 MR. NOONAN:
I'll repeat it.
s
f
' i'(4 e
o-179 v+
. ~. - ~..
s 1
We'll look at the mechanics or. the process 2
for extracting selected items for sampling.
3 What we propose to do here is have the staff p
4 onsite, and then we come back one month from today--not from 5
today, but one month,'in a public meeting and report on all 6
of the staff work to date, what we did.
7 I basically established a procedure on these t
8 type of audits, the staff will not discuss with you until 9
the results and conclusions.
That will be done in a public 10 meeting.
's
/
II q[qf j
John, I was also asked, from a regional
.t.
12 standpoint--and I'll read this also--
We believe that the 13 construction adequacy program should include the verification 14 of the compliance of procured materials, and items with 15 procurement / design requirements.
I6 I think that was a question that was asked 17 of you.
You basically said you were not going to do that, I8 and the staff is not satisfied with that particular answer.
19 So there will be discussions of that subject also.
20 One other item that I would like to point s
?-
21 out is basically the--and, Bill, I guess this is one really r
22 directed at you--the procedures that we're now looking at
' ' 23 are in draft form, and I made the statement earlier that t4 24 you proceed at risk on these kind of procedures, but from 25 the viewpoint of the. staff, including the regional people, i-
180 I
feel that it's to the point now where it's getting--we're 2
looking at too much drafts, and we want to be able to 3
actually take possession of the procedures and look at them.
4 And I've instructed them they cannot do that; 5
if they do that, it's going to go to the public document 6
room.
7 I guess the bottom line is that the drafts--
I they're waiting'for the final product, and it's important I
that we now start seeing that.
10 one other item, I, guess, and since we'll be 11 coming back in a month to do the staff report, I would like 12 also to ask you to consider starting your monthly meetings 13 on the progress of work.
I4
__I would like to do that, on the monthly meetinga, 15 establish a time frame so we can anticipate it from month 16 to month.
It doesn't have.to be a date, but at 1, east some 17 time frame we can plan on.
18 Any questions?
MR. COUNSIL:
Vince, since this next monthly 20 time frame, we've got at least three weeks of audits, it 21 sounds like, in various locations, I think that we can agree 22 now that we would hold a monthly meeting, the first one in 23 conjunction with your public meeting for your audit summary, 24 and if those audit summaries continue in the same type 25 fashion, we'll just tail-end right' along with you.
c
.o 181 1
MR. NOONAN:
Okay.
2 MR. COUNSIL:
I.think it will be~ easier that 3
way for your planning, our planning, to go a day and a half 4
like we did this time, or whichever, and have an audit s
3 summary in addition to where we are in our monthly meeting.
6 MR. NOONAN:.Okay.
7 What I propose is to minimize the amount of
[3 travel for the staff, particularly coming down here; would 9
be maybe the first day we could have the half-day session, 10 we can fly down in the morning and start at 1 o' clock and 11 have the staff go through their results, and then the next 12 day, I would want to plan for a full day, to have you come 13 along and give us the status of your work.
14 MR. COUNSIL:
That's fine.
I think we can 15 also then go on that kind of planning, work on a generic
~
16 type of aggenda.
17 If that's all right with you I'll work with 18 Charlie--
19 MR. NOONAN:
I think that's--
20 MR. COUNSIL:
Also, I think for the next 21 couple of months he and I can come up with not only a generic 22 aggenda,'but tentative dates, so that everybody in the room 23 can plan on those tentative dates and can be advised in a 24 public notice.
25 MR. NOONAN:
Okay.
l "t
4 4
-b :
182 1
l'll have Charlie work with Ms. Ellis and 2
k.eep her informed as to times and dates of the meetings, 3
so forth.
4 More likely, I'd like to keep this area right 5
here, if we could.
The room is really large enough, seems 6
to be adequate.
7 I guess with that, unless there are any 8
other staff comments or anything from the Utility, I'd 9
like to go ahead and close the meeting.
10
[No response.]
11 MR. NOONAN:
Thank you.
12
[Whereup'on, at 2:15 p.m.,
the meeting was
(
13 concluded.]
s 14 15 16 17 18 19 20 21 22 23 24 25
.