ML20073R749

Affidavit of Ma Walsh Re Computer Program in J Doyle 830426 Surrebuttal Testimony.Strudl Program Utilized W/Sufficient Similarities to Assure Results Will Be Accurate

ML20073R749
Person / Time
Site:	Comanche Peak
Issue date:	04/26/1983
From:	Mary Walsh Citizens Association for Sound Energy
To:
References
NUDOCS 8305040529
Download: ML20073R749 (1)

Text

,. _ ._ _ ___ _._ .. _ __ _ _ _ _ ____ . _ _-

s I

4/26/83 UNITED STATES OF AMERICA .

NUCLEAR REGULATORY COMMISSION "^ M BEFORE THE ATOMIC SAFETY AND LIC'ENSISG BOARD

_4 .'83 R -2 A10 :48 -

In the Matter of ,

g APPLICATION OF TEXAS UTILITIES I ~

Dodkht.NhMi50-445 i GE&IERATING COMPANY, ET AL. FOR -

. ' BiiiiYdi S 0-446 i . AN OPERATING LICENSE FOR _

g . . . _ .

COMANCHE PEAK STEAM ELEC.TRIC STATION UNITS #1 AND #2 - I '

- I I +

(CPSES)

[

AFFIDAVIT OF MARK ANTH0NY WALSH

~. '

l ,

My name is Mark Anthony Walsh. AttherepuestofJackDoyle,I.ran

the computer program which is attached as CASE* Ekhibit 761 A to Mr. Doyle's

. ~

4/26/83 Surrebuttd1 Testimony. - -

Although the' computer program which I used is not exactly like the

~

STRUDL program used at Comanche Peak, it is a STillIGL program with sufficient similarities to assure that the results will.be accurate.

This affidavit wa's prepared under my pers'enal' direction, and the thoughts and words, expressed h'erein are my own thoughts and-words (with the exception of minor grammatical changes, either to correct.sp,elling or to. clarify my

~ ~ ~~

4 meaning, which did not chinge the intent of my thoughts)^. ~~~

I N)

Mark Anthony Walsh /

Daie

4/26/83 STATE OF TEXAS r - - - - ~-

{ ~ -

On -this, the 26th day of April,1983r pers'onally appeared Mark Anthony Walsh, known to me to be the person whose name is subscribed to the foregoing -

instrument, and acknowledged to nie that he^ executed the same for tie l purposes herein expressed. ' '

Subscribed and sworn before me on the 26th _ day of [pril,1983. s.

m ?w

.. Not'ary Public in nd for .the State ofiTexas

~

MJp/fj-

~

My Commission Expires: . l

/ / s 8305040529 B30426 -

30 3 PDR ADOCK 05000445 '

x 1 G PDR

o

. CASE EXHIBIT 761 UNITED STATES OF AM' ERICA 4726/83 NUCLEAR REGULATORY COMMISSION Sr te:Mii t BEFORE THE ATOMIC SAFETY AND LICENSING BOARD 4 .g3 Yad ~7 gg %8 In the Matter of .

I - -'

I- - x .'

APPLICATION OF TEXAS OTILITIES  ! ~

GENERATING COMPANY, ET AL. FOR I Docke'tI Nos.[50-445 k ~and 50-446 AN OPERATING' LICENSE FOR .

' COMANCHE-pears STEAM ELECTRIC. . I

7- -

STATION UNITS #1 AND #2 I (CPSES) -

I~ '

1 -

+

SURREBUTTAL TESTIMONY OF, JACK D0YLE,

, WITNESS FOR INTERVENOR CASE r . - - - -

1:

1 Please state your naine and address for} t e record.

2 A: ' My name is Jack Doyle. My address is.P.40. Box 64, Turnpike t - Station, Shrewsbury,-Massachusetts' 01545. e-- - -- -

4 Q: Are you the same Jack D.oyle who testif.ied as a witness for CASE _

5 in the September 1982 operating license hearings for Comanche Peak? -

~ '

6 - A: Yes, I am. ,

M. .[.

~

7 . Q: In preparation of your testimony for the May 16-20, 1983, operating 8 license' hearings for Comanche Peak, was there an item of particular concern s 9 whjch you felt should be, sent iimledjately to a'11' parties?_ . .

10 . A: Yes, there was. -

Please tell us what that item is and what your co'ncer5s ar.e

~

11 Q:

12 regarding it. - _

13 A: The ' problem with which I am concerned was discussed somewhat in 14 CASE's 4/20/83 Brief Regarding C6nsideration of LOCA in Design Criter.ia

- - - . - - . -  ;- . ~

15 for Pipe Supports (beginning on'page 33, secony paragraph, through page .

~

16 34, second full paragraph). In addition, I will be discussing'Applic' ants' i

i I '%

~

._ ~

.- : . . ~ -

. =.

l

~

• 0h ^ r , , n ~

^

pe n : 5 -

l

1 Exhibit 1420 regarding LOCA temperature effects on pipe supports.

2 Q: Why do you feel that this matter is so import

• ant that it should 3 be gotten into the hands of the Board and all parti,es immediately? .

4 A: The conclusions of the NRC Staff and the Applicants relative

5. to self-relief of LOCA effects at.th'e joints 'is; bas @d on this material, 6 which the NRC Special Inspection Team reviewed-during its investigation 7 of the concerns of Mr. 'Walsh .and me. T.be accuracy of this material 8 .must be assured. Iri the case of the equations and formulas in question, 9 ~ } their, accuracy is negated by an erroneous apprIIactr. There is n'o assurance 10 that the supports will functi'on during a LOCA cr immediately thereafter.

• 11

• A further question arises:- If these squations. are in error, -

-12 what other equations that are used an proof of a position are also in

~

~

13 error? These equations.are prepared, checked, reviewed, - - -

and approved.

~

14 They were further blessed. by the ~ NRC Specia.1 -Inspe.c. tion Team. -

15 Q: Please tell us what your specific c6ncerns are regarding this 16 matter., -

17 A: .In the case of the moment ' restraint structural support, problem

~~ ~

' ~ ~

18 450 (CASE Exhibit 761B), in determining the ' structural stiffness at joints 1.9 1, 4, and 5, the stiffness of only one anchor bolt wasi consider _ed. In

~

20 fact, there are multiple anchor bolts, and of more consequence, shear keys are welded to the base plate whichiesults in k factors several

~

21 22 hundred times those used by Gibbs, and Hill, who apparently preparsd the -

23 - e calculation. .

- ~ '

~

24 These items, the anchor bolts and shear keys, are a paraQel.sp. ring 25- system. Effectively, these joints will act as rigid connections unless -

• e b g -O y O b

6 *%

9 w

N

~

1 there is a catastrophic fairlure of the concrete in the area of the 2 shear keys, which will result in unpredictable consequences. For more 3 details regarding this, see CASE Exhibit 761C. ~ -

4 Regarding the upper lateral restraint (CASE Exhibit 758, referred to

-+ . .

5 on page 33, second paragraph, of CASE's 4/20/839 Bri@E on LOCA) this is 6 approximately a 9 ton beam located a considerable distance above critical,

~ ~

7 piping in the steam generator _ area. The_ analysis' for this beam considering 8' the flexibil-ity of the walls is based on a simple three-springs series.

~

9 , Gibbs, and Hill, in their approach, used an obifque approach tha't failed .

10 to consider, among other thin ~gs, several critical factors.

11 . (l') The beam reacts its.-Toads to a vertical slab, not a beam. The -

12 method for determining the stiffness of the wails, therefore, should incor-

~

13 porate either the " Marsh method,1904 cross-beam-process,"- the "0kuna -

14 ' Abrams anologous grid inethod, 3952," or show-that dimensionally the _..

15 above procedures are not required for the aspect ratio in question.

~~

16 Further, the'transf5rmed section shou'ld have"been used to determine the .

17 moment of inertia "ACI cracked beam procedure".1 0 stead of the monolithic

~ ~

~ ~~ ~ ' ~ ~~

18 method used. --

19 (2) Since the wall deflection is ansiular to the later.al beam, a

20. compor,ent to the deflection induces a side load into the beam across' 21 its minor axis, which was not considered-22 (3) In the interaction: equation in _the analysis of the upper-lateral 23 - restraint,-Gibbs'and Hill'us'e' a d compress-ion allowable in excess of' 2 24 yield for A50 steel in-a 280 F0 e.nvironment.  : _

/

25 As a result of these errors a6'd other funfamental errors in the .

.~ : .. ' ~ [

~

a es 4

, g.  %

N

S 1 equations themselves, I decided to. approach the problem independently, 2 using, however, the k factors of the walls, area of t% beam, Young's 3 modulus of the concrete, and. coefficient of expansi.on and dimensions 4

indicated in Gibbs -

and Hill's calculations. The results of approaching

5. the problem from a springs-in-seri_es' position rpsults .in s.ignificant 6 increase in the loads in th'e beam and on-the wall. The hand calculations -

.s.

7 as shown in CASE Exhibit 761C on the lateral beam were subjected to 'a 8 computer analysis substituting members of equivalent spring rates to those

~

of the wall.

~

9

• ~~~

10

~

Q: And what were the re'sults of those calcutations?

. N .

11.o A:' The results of the hand calculations-supported .by the computer - -

12 analysis indicate a gross error in the fundamer" Ital approach as used

~ ~

13 in the Gibbs and Hill calculations. (See CASE.' Exhibit 761A for the

~14 computer analysis.) .,

~~

15 Q: Who perfonded the computer analysis 5 '

16 A: It was perfonned by Mark Walsh at my request.

17 Q: Do you have anything further? ,,

~

18 A: In reference to Applicants' Exhibit 1420, three equations listed

~

~

19 as Case 1, Case 2, and Case-3 were suppiled to indicate the self-relieving _

20 characteristics of Hilti anchors, Richmond anchors, and Nelson studs.

- - ~

21 In all three c~ases, the procedures are in error. These are again a problem 22 of several springs in series and additionally,;in the case of Hilti

' ' ' ~

23 Kwik-Bolts-and Nelson studs,' the joints are a parallel ' system of multiple 24 springs. For the case of Hilti anchor balts aiid Richmond anchors., . ,

25 see material contained in CASE Exhibit 761C. AsforCase-3,glsons a y o

~

~

1 .

x

5- .

I studs, the Gibbs 'and Hill calculations neglect a critical component 2 in the effective spring system. These are , embedded plates,and therefore w

3 shear is transferred.through.the concrete and the Nelson studs in parallel:

4 . The effects of Jhe concrete have not been considered by Gibbs.and Hill e4 * '

5- and theref' ore their approach would- be in gross ierror'7 '

• 6 Q: What does all this inean? ~_ .

7 A: The Applicants, Gibbs and Hillr and the NRC have predicated ,

8 .

their arguments that thermal expansion during:.a LOCA or in fact during g nonna.1 plant operating conditions is of no consequence in the design

~ ~

10 calculations, since the self-relieving charac.teristics '

of the joints

. = -

~

114 ,_ pteclude overstressing.' This"is bas ~ed on fiv3[~c.alcula.tions.whi.ch were ,

12 supplied to . CASE during discovery and in Appkicants Exhibit 1420. These 13 contain gross ' error in fundamental engineeririg 'oFinciples.' __

~

14 _ The question ~arisesrWhere are the checks' and balances system -

15 which assures .the safety of a nuclear power plant if such gross errors 16 can be introduced, checked, reviewed, approved, and blessed by the NRC?

17 How far does this acceptance of gross error extend into other areas to 18 which we are not privy, where 'we are told that 'something is acceptable 19 because a. document exists, but 'where the_ credibility of the do.cument 20 . it.self is now .in doubt? What assurance.do we have ,that the checks and 21 balance system has not broken down, not only at Comanche Peak, but in

_ . , - ~

22 the nuclear industry generally?_.

~

~' '

. a. _

~. . .

a - - . _ . ~

v .

j

• m S "

m g A e *g b

I. .

x

1

.* j I have read the foregoing 5-page Surrebutta.1 Testimony of Jack Doyle, );

4 ,

Witness for Intervengr Case (CASE Exhibjt 761), wh,ich was, prepared under j w

my personal direction; the thoughts and words expressed therein are my

,pwn thoughts and words (with the exception of minor grammatical changes, either to correct spelling or to tlarify what } meant, which did not change the intent of my thoughts). ~ Where quastions w'ere .gosed, they were posed by CASE. This testimony is true and currect,to the best of my knowledge

. and belief. . . 1.

~

A -

Jpck D,0yld p Date: 4/26/83.

~

STATE OF TEXAS '

On this, the 26th day of April,1983, personally appeared Jack Doyle, known to me .to be the person whose name is s.ubsc,ribed to the foregoing instrument, and acknowledged to me that he executed the same for the purposes therein-expmssed. , - - - -

Subscribed and sworn befo,re me on the 26th day of April, 1983.

' ~

J.

. Not&rst_gblic P in a the State of Texas

.- My Commission

~

Expires: O /I~~ . -

-/

l 2 .

= . I

- ~..

4 em "

=

e g

> f 4

• a I