ML20074A545

Supplementary Surrebuttal Testimony of J Doyle.Error in Fundamentals Used in Calculations for Upper Lateral Restraint Compounded by Errors in Stiffness Used in Gibbs & Hill Calculations.Related Correspondence

ML20074A545
Person / Time
Site:	Comanche Peak
Issue date:	05/09/1983
From:	Doyle J Citizens Association for Sound Energy
To:
References
NUDOCS 8305130157
Download: ML20074A545 (2)

Text

wer;c comchini il CASE EXHIBIT 805 5/9/83 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISS10rl , r ,

g BEFORE THE ATOMIC SAFETY AllD LICEN CQARD 9 MQ' In the Matter of I .b iB I

C Afgy 1 ---

I 2 79 APPLICATION OF TEXAS UTILITIES No !h) ,f5

'. GENERATING COMPANY, ET AL. FOR

.AN OPERATING LICENSE FOR I

I Docte, 4' Tig - hs d 5 d46

' \-

COMANCHE PEAK STEAM ELECTRIC I #

STATION UNITS #1 AND #2 I (CPSES) I L'

i

. SUPPLEMENTARY SURREBUTTAL TESTIMONY OF JACK 00YLE, WITNESS FOR INTERVENOR CASE 1 Q: Mr. Doyle, is there additional surrebuttal testimony which you 2 wish to file regarding the document received by CASE today which has been 3 marked as CASE Exhibit 790, a drawing of "R. B. Internal Struct. Equip.

4 Supports & Fdns., Sh. 1, 2323-SI-05507 5 A: Yes, there is. In doing the analysis on the Gibbs and Hill

~

6 calculations for the upper lateral restraint (see CASE Exhibit 761C),

7 I used the same numbers as Gibbs and Hill sheet Nos. 32 and 33, Case 1 8 and 2, for walls "A" and "B" (see CASE Exhibit 758). However, on receipt 9 of th'e letter from Applicants' counsel dated May 6,1983, with attachment 10 (drawing of upper lateral restraint - see CASE Exhibit 790 attached), I 11 find that I must revise my estimate of the problem.

12 Ine centerline of the ul i per lateral restraint is 858:6" '(see Section 13 14-14, Drawing S1-0550, CASE Exh' bit 790). The location'of the upper 14 lateral restraint relative to the steam generator may be found on Figures 15 1.2-8,1.2-13,1.2-14,1.2-18,1.2-19 and 1.2-20 of Section 1.2 of Appli- -

16 . cants' FSAR (Applicants Exhibit 3 - we are attaching copies as CASE Exhibits S305130157 830509 I PDR T

ADOCK 05000445 '

pop

e .

I 1 799, 800, 801, 802, 803, and 804, respectively, for the cravenience of the 2 Board). From these drawings, the stiffness of walls "A" and "B" are much 3 higher than was shown on the calculations.

4 -

For example, in the case of wall "A," the calculations show 42 inches

~

5 as the wall thickness. From the cross-sectional drawing listed above, this 6 appears to be less than half the thickness that actually exists (about 8 7 times the stiffness'used by Gibbs and Hill). Beyond this, the point of 8 connection for the lateral restraint beam (on the reactor cavity wall) 9 is beneath the missile shield. Any failure on the far side of the wall 10 could result in missiles falling on top of the reactor head about 30 feet 11 below.

12 Wall "B" offers even more interesting information of differences 13 oetween the Gibbs and Hill numbers and the numbers to be found in the real 14 world. Wall "B" for Case I indicates the wall thickness to be 45 inches 15 and Case II indicates the thickness as 33 inches. The cross section at 16 elevation 860'0" shows the beam to be reacting directly into the floor 17 slab'and at elevation 861'4" (the opposite side) shows the beam to be 18 reacting partially into the major platform slab (due to stiffness differentials 19 between the wall and slab, this is effectively as if the load were totally 20 reacted by the platform = lab).

.21 The error in the fundamentals used in the calcs is now compounded 22 by the errors in stiffness as used in the calcs by Gibbs and Hill. I cannot 23 clari_fy this particular error due to a lack of drawing which would provide 24 the dimensions required to determin'e the actual stiffnesses of thc various 25 walls. There is an additional problem which lacks sufficient information 26 for elaboration and that is the existence of walls perpendicular to wall 27 "B" in proximity to the reaction point for the upper lateral restraints.