ML20093M815

Affidavit of J Doyle in Form of Partial Response to Applicant Statement of Matl Facts as to Which There Is No Genuine Issue Concerning Applicant QA Program for Design of Piping & Pipe Supports

ML20093M815
Person / Time
Site:	Comanche Peak
Issue date:	10/18/1984
From:	Doyle J Citizens Association for Sound Energy, TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC)
To:
Shared Package
ML20093M798	List: ML20093M796 ML20093M804 ML20093M815 ML20093M838
References
OL, NUDOCS 8410230308
Download: ML20093M815 (9)

Text

-

UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter . of i I

TEXAS UTILITIES GENERATING -

1 Docket Nos. 50-445-1

> COMPANY, et al. 1 and 50-446-1 I

(Comanche Peak Steam Electric Station i Station, Units 1 and 2) {

. CASE'S PARTIAL ANSWER TO APPLICANTS'. STATEMENT OF MATERIAL FACTS AS TO WHICH THERE IS NO GENUINE ISSUE REGARDING APPLICANTS' QUALITY ASSURANCE PROGRAM FOR DESIGN OF PIPING AND PIPE SUPPORTS FOR COMANCHE PEAK STEAM ELECTRIC STATION in the form of AFFIDAVIT OF CASE WITNESS JACK D0YLE 0: Do you have any comment on Applicants' Statement 1, which states:

"Each'of the responsible design organizations for piping and supports at Comanche Peak has established procedures to implement the provisions cof 10 C.F.R. Part 50, Appendix B and ANSI N45.2.11 to their respective applicable functions in the piping and support design process.

(Affidavit Table IV.1)"

A:' No -- I am not as concerned with whether a procedure exists as with the fact as to whether.or not it is being implemented.

Q: Do you have any comment on Applicants' Statement 2, which states:

" Regulatory' Requirements and licensing commitments set forth in the license application are incorporated into design specifications by

.Gibbs & Hill for Comanche Peak for both piping (Class 2 & 3) and supports. These specifications are transmitted to the responsible design organizations for incorporation in their design process.

Similarly, Westinghouse has established-a specification for the design of Class 1 (and Class 1 extension) piping. Westinghouse employed the Gibbs & Hill specification in its design of non-Class 1 auxiliary piping. (Affidavit at 16 (G&H), 25-26(U).)"

A: No, for the same reason as in my answer to Statement 1 preceding.

8410230308 841018 i PDR ADOCK 05000445 O PDR

= - - -

wz.Q;c, -

s ~J LQ: .Dolyou have anyLcomment on Applicants' Statement 3, which states:

"Each of:the pipe support design organizations has incorporated the Gibb's &.HillLapecification applicable to the' design of pipe supports

'into their design process. This-specification is incorporated into eachlorganization's designs (including drawings, procedures, instructions and~ guidelines:as appropriate) in accordance with establishediprocedures. (Affidavit at 32-33 (NPS), 39.(ITTG) and 43-44

?(PSE).)"

~

^

A:. .No I-do not -- for-the same reasons as for answers to Statements 1 and 2'; preceding.-

10:? ':Do' you-have~ any comment int Applicants' Statement 4, which states:

" ~

Each: design organization has' implemented design control measures which einclude., verification and/or checking of the adequacy of each design,

including the. initial design.of the piping or support prior to release-of the design-for construction.' These measures include documentation of;the design reviewer's findings and correction of'the deficiencies by

the original designer. . Each design organization also requires that the person performing design review may not be the same person who performed the original design, although he may be part of the same

. organization as' the original designer. (Affidavit at 20-22 (G&H), 30 (W),135-37 (NPSS), 40-41-(ITTG), and 46-48 (PSE).)"

cA: LY es, I:do.- The statement is made that Applicants have measures to find and correct ~ deficiencies by the original. designer.

!The Cygna Phases.1, 2, and:3 independent review /1/.shows for

.' Phases 1 and 2 that of 9 vendor certified' calculations, 6 contained l gross errors,f2/; for: Phase 3,18 of 22 main steam supports contained

gross engineering errors (see CASE's First Motion at page 49); of the

-66 component cooling' support calculations, 70 per cent contained errors

'(see CASE's First Motion at page 73). ' Of the 97 calculations contained

/1/ See Independent Assessment Program for Comanche Peak Steam Electric

' Station, Prepared for' Texas Utilities Services, Inc., Prepared by Cygna Energy Services: Phases 1 and 2, Draft Final Report, November 5, 1983; Phase 3, Final Report, July 16, 1984.

/2/ See' CASE's'First Motion for Summary Disposition Regarding Certain Aspects.of the Implementation of Applicants' Design and QA/QC for

Design (hereinafter referred to as CASE's First Motion) -- at page 49 i

2 t

L

7 '

in Cygna Phases 1, 2, and 3 review,.at least some (4) had to be repaired to prevent structural failures, not including the generic fixes originally mentioned in Walsh/Doyle (see CASE's First Motion:

page 73, answer 14; page 76, item 6; and page 77, item 8).

Q: Do you wish to comment on Applicants' Statement 5, which states:

"During the course'of construction of the piping and support system changes in design of supports are virtually unavoidable.

Implementation of the changes are governed by established procedures

-and instructions. The most commonly employed method to implement such changes is through Component Modification Cards ("CMCs"). These changes are' subject to design review, verification and approval in accordance with procedures commensurate with the design review process employed in the original design. With respect to design changes not initiated by field modi $1 cations, each organization also conducts design reviews of the change in a manner commensurate with the procedures for new designs. The design change control process for each organization provides^that the organization which performed the original design to also perform the design review of the design changes. (Affidavit at 50-56.)"

'A: No, I would prefer to summarize relative to the contents of Applicants' Affidavit relative to the contents of Applicants' Affidavit in cupport of their Motion.

The reason is to avoid a lengthy and redundant cruise through the-maze of deception by Applicants. The question is simple: Did the Applicants have a QA/QC program in place and, if so, did they implement ,

it?

The answer to the first question is: The Applicants may have had

.a program, as can be noted from the following:

Applicants and their agents again and again, collectively and individually, assure us that a multi-level checking process existed and that this process was backed up with a critical / independent audit system (see Applicants' Affidavit: Gibbs & Hill at pages 21-23; Westinghouse at pages 30 and 31; NPSI, pages 34 to 36, ITT Grinnell, 3

+

pages 39 and 40; PSE at pages 44 to 49).

Applicants even go beyond this and supply two examples of how the system worked in detecting errors (see pages 73 and 74 of Applicants' Affidavit).. At page 77, Applicants extend their examples to the guidelines (Section'XI Welding) and how it worked there.

Applicants go on to assure us that recurring errors are caught and resolved by their iterative program (see page 72 of Applicants' Affidavit)..-But this does not conform to reality, since recurring errors are rampant in CPSES calculations. See, for example: weld calculational. errors (4); calculations for structural components missing (4); no calculations for snubbers (4); wrong section properties used in calculation (3); to name a few of the repetitive er'rors found

~

by Cygna in only 22 calculations (see CASE's First Motion at pages 49 and 50). .

'Q: 'Are there other areas which indicate Applicants' inability to turn out

. proper' calculations?

A: Aside from the vendor certified drawings, we have.in the past had the unique experience of finding calculational errors in Applicants'

^ '

procedures which, strangely, were always in Applicants' favor. (Of these 6, items 3 and 6 in the following examples were, however, from a vendor certified group; the remaining foru were prior to vendor certification.) Errors in Applicants' favor:

(1) From CASE Exhibit 669B f3/, items 4G and 4H: support CC 028-039-S33R, item 6 was a 6" wide flange that had web

. /3/ See CASE Exhibit 669B, Attachment to Deposition / Testimony of CASE Witness Jack Doyle, accepted into record at Tr. 3630.

4

bending problems'and this is the one which NRC Staff Witness Dr. Rajan had to back off in his Affidavit; Applicants were wrong in their analysis, and Dr. Ra'jan was wrong on all of his assumptions, andlth'at one was rebuilt /4/.

(2) From CASE Exhibit 669B, items 11GG and 11HH: support CC 008-709-A43K, item 5 was a 16" diameter diaphragm, and here G again Applicants were wrong in their analysis, and the initial analysis by the NRC SlT team was erroneous. The the SIT noticed an error in. Applicants equation and a re-evaluation showed that that plate failed, and it had to be modified.

(3) From CASE Exhibit 669B, item 11TT: support CC-1-107-008-A23R, item 1, Applicants failed to include the torsional stif fness of item 1 and performed'a finite element analysis and a physical test to prove the part was marginal but 0.K.

However, when the SIT requested a reanalysis using the actual stiffness, it turned out that the loads on this particular support increased 660%, which made the marginal deflection problem academic, and the support had to be rebuilt. -

(4) CASE Exhibit 6.69B, items 11NN and 11-00: support CT-1-008-006-S62K, item 9, a 3/8" plate, was underdesigned. I don't recall ~anyone's addressing that problem other than us.

(5) On the upper lateral restraint, Applicants made a gross error in the stiffness calculations which caused a reduction in the f4/ See 12/13/83 Affidavit of Jai Raj N. Rajan on Support No. CC-1-028-039-S33R, attached to NRC Staff Motion to Reopen Record to Admit the Affidavit of ' Dr. Jai Raj N. Rajan.

5

5fg - u;

~

t . 1

, le ?

c, :.-

i

. thermal loadings of significant proportions, an'd by

~

~ ? Applicants' equations, the support just 'oarely passed. But for--future analysis, to show the adequacy of the upper -

latera1Lrestraint,' Applicants had to resort.to a different

~

mathematical' procedure than originally used in.an effort to

show'that the support was 0.K.

(6)- InDaddition,-in' Applicants analysis of support SI-1-325-002-S32R,- they used an erroneous assumption to decermine the thermal gradients (plus other erroneous calculational

~

4

. assumptions). .The result'was that this error allowed Applicants.to show'the support as having astress ratio of .9, W 'whereas in reality the support would fall. f5/

'With the preced'ing and paragraph 2 of my answer"to Applicants

Statement 4. -it is : obvious that regardless of what-program Applicants

~

- ha'd-in place,-it was not' implemented.-

s

.Q: Do you wish to answer any points in the affidavit accompanying Applicants' Motion which do not correspond to an item in the Applicants' Statement'of Material Facts?

- A: <Yes,.I do. - On page 15 of' the : Affidavit, Applicants quote from an '

article'which~ appeared in " Power" Magazine, February 1979 (Attachment D to Applicants' Affidavit), and adopt the statement which reads as follows:-

25/ See Attachment A to Affidavit of Applicants' Witness.Finneran, attached ito Applicants' original Motion for Summary Disposition Regarding Local Displacements and. Stresses; see also discussion in CASE's Answer to

' Applicants' Reply to. CASE's Answer to Applicants' Motion for Summary 1 Disposition-Regarding Local Displacements and Stresses'in the form of Affidavit of CASE Witness Jack Doyle.

6 n --

c --.

a "The' hanger-design process is not simple._ It is compicx and tedious, involving many disciplines at the A/E firm, at the hanger manufacuring plant, and at the site. The process is iterative, 1 continuing until the plant goes operational. (Emphasis added.)"

I find two problems with this attempt by Applicants to justify their methodology: (1) The article, which is taken out of context, does-not imply that the iterative process extends equally to construction. The article is addressing an iteration of design and layout before' construction and problems which are most often the result of s) ace' allocation and interferences between the various disciplines.

(2) The article in no way implies-that, since the process is iterative, errors'in engineering fundamentals are therefore acceptable until the

~

completion of the1 final iteration.

The article, which'is describing design processes whereas Applicants are involved in construction iterztion, and in many cases, the resolution of these problems was indicated by and the result of Walsh/Doyle and not the result of some master plan by Applicants.

While I and most other engineers accept the design iterative process for nuclear, petrochemical, as well as aerospace and other

~

industries,.the' idea of construction iteration is indicative of a failure to utilize the design iterative process.in the first instance.

For; example, the design iterative process should be utilized to reveal interferences that may occur in the field by inter-disciplinary review of ' preliminary designs for all disciplines prior to fabrication and construction. The field is not the arena to uncover interferences.

Q: Do you have any further comments?

7

, _ . _ .... . . - - . . __. . . _ _. ____ m . . . . . _.

Wi ~ .+- -

i. -

a q.

~, . . ,

y.. ,

i..

IA: - No, because the above precludes the necessity to answer any of.the-2 o

I- ,

other items listed by Applicants. Once a program has b3en exposed as a

-total failure,'~why beat a dead horse?

i

?-

h 4

m E.

~_A 4'

s i'

s 4

J 6 -

t i -

u i

N 4

1-e f.

b.

f;

[._ s

i. ,

[-

.. o i b

e 8

y9 6-* M w .m-ym .d -we s .-.4 -wy - s- N-'Nq%-yw$d-'Ci--19mP w*ht -+ m ' iy 4 e 't-We'~ INT-mf 4 Ww te *wwMP*'Y *rweWN-~TNm* *+*v*

x s

I have read the foregoing affidavit, which was prepared under my personal direction, and it is'true and correct to the best of my knowledge and belief.

V /

Date: ,

h /f STATE OF I%

COUNTY OF- MN ub On this, the 16 t(day of bkba ,1981, personally appeared 4

\.

0 hmk.0

, known to me to be the person whose name is subscribed to the foregoing instrument, and acknowledged to me that he/she executed the same for the purposes therein expressed.

Subscribed and sworn before me on the io N day of dh~

~

1981 -

rn s- 0- c Notary Public in and for t State of hk My Commission Expires:

-_