ML20079Q069
| ML20079Q069 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 05/04/1983 |
| From: | Doyle J Citizens Association for Sound Energy |
| To: | |
| References | |
| NUDOCS 8305110354 | |
| Download: ML20079Q069 (42) | |
Text
__-. . - - - . - _ _ _ - _ - -. .-
IL NM'57 F AMERICA CASE EXHIBIT 763 Y COMMISSION
, NUCLEAR REG
,. -- ,5/4
, / 83.s BEFORE THE ATOMIC SAFETY AND LICENSING BOARD O\';$
, .? '
In the Matter of - '
. r.
APPLICATION OF TEXAS UTILITIES !
g Docke't;Nos. 50-4457.
GENERATING COMPANY, ET AL. FOR , o. ; 1a nd 50-4 46-AN OPERATING LICENSE FOR f,, ..: ' 4%-
COMANCHE PEAK STEAM ELECTRIC / t' STATION UNITS #1 AND #2 Q p/ im '. '_ 4, . .
(CPSES) g TABLE OF CONTENTS SUPPLEMENTARY SURREBUTTAL TESTIMONY OF JACK D0YLE, WITNESS FOR INTERVENOR CASE Page ITEMS CONTAINED IN SIT REPORT:
4 Item c. Design Analyses of Richmond Inserts and Hilti Bolts 1 .
Item d. Differential Themal Expansion Effects in Pipe Supports 5 Item'e. Differential Themal Expansion Effects in Wall-to-Wall, Floor-to-Ceiling, and Floor-to-Wall Pipe Supports 6 Item f. Stability of Pipe Supports Designed for CPSES 9 Item g. Use of U-Bolts in Pipe Support Design 10 Item h. Loading Due to the Seismic Acceleration of the Pipe Support Structure 16 Item i. Moment Restraints and Local Pipe Stress Due to Welded Stanchions on Pipes 18
- Item j. Deflections and Local Stresses in Pipe Support Structures 19 Item 1. Consideration of Kick-Loads 22 n a. '
EEE Item p. Welded Stepped Connections, Fillet Welds and Skewed go gf Welds 24
! g) Item q. Section Property Values Utilized by PSE 27 no Item r. Support Pads Welded Over Pipe Girth Welds 28 m
8 h
coa.s Item s. Damage, to Pipe Support During Hydrostatic Testing
. 29
L TABLE OF CONTENTS Page 2 ITEMS CONTAINED IN SIT REPORT (continued): Page Item 4. Inspection of Vendor Certified Supports 30 Items a, b, k, m, n, o 37 ITEMS ADDRESSED BY WALSH/00YLE BUT NOT IN SIT REPORT 31 NRC STAFF DID NOT CONTACT MR. D0YLE REGARDING HIS CONCERNS 35 ADDITIONAL QUESTIONS REGARDING LOCA 36 SIT REPORT
SUMMARY
AND CONCLUSIONS. ETC. 37 ATTACHMENTS TO 5/4/83 SUPPLEMENTARY SURREBUTTAL TESTIMONY:
CASE Exhibit 716 - PSE Guidelines, portions of Section XI, Weld Calculations CASE Exhibit 763A - FFTF REPORT, DRILLED-IN EXPANSION BOLTS UNDER STATIC AND ALTERNATING LOAD CASE Exhibit 763B - RESPONSE TO NRC - NRC Site Meeting -
January 18, 1983 CASE Exhibit 763C-1 through 763C Component Modification Card (CMC) No. 46730 and Revisions CASE Exhibit 763D-1 through 763D-4, 763D-6 through 763D Component Modification Card (CMC) No.
46174 and Revisions CASE Exhibit 763E - ITT Grinnell Catalog PH79, Pipe Hangers, pages ph-20 and ph-21 CASE Exhibit 763F - STEEL STRUCTURES Design and Behavior, by Salmon and Johnson, Structural Fasteners, pages 84-87 CASE Exhibit 763G - MANUAL OF STEEL CONSTRUCTION, SEVENTH EDITION, American Institute of Steel Construction (AISC) .
CASE Exhibit 763H - Support CC-1-107-008-E23R Calculations CASE Exhibit 7631 - Portions of Cold Form Institute steel manual used at Comanche Peak by at least one group I
j CASE EXHIBIT 763 5/4/83 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION I
BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of I
APPLICATION OF TEXAS UTILITIES Docket Nos. 50-445 GENERATING COMPANY, ET AL. FOR I and 50-446 AN OPERATING LICENSE FOR I COMANCHE PEAK STEAM ELECTRIC I STATION UNITS #1 AND #2 (CPSES)
. SUPPLEMENTARY SURREBUTTAL TESTIMONY OF
_ JACK 00YLE, WITNESS FOR INTERVENOR CASE 1 Q: Mr. Doyle, is there additional testimony which you would like to 2 present to the Licensing Board in regard to the concerns raised by Mark 3 Walsh and you and the investigation of those concerns by the NRC Special 4 Inspection Team (SIT)?
5 A: Yes, there is. My testimony will deal primarily with the Special 6 inspection Team's Inspection Report 50-445/82-26,50-446/82-14(hereinafter 7 referred to as the SIT Report).
8 Q: What is the first item you would like to address?
9 Item c. Design Analyses of Richmond Inserts and Hilti Bolts 10 A: The first item is item c. of the SIT Report (beginning on page 16).
11 First I must state that in addition to LOCA (loss-of-coolant accidents),
12 normal operating temperature differentials are also a concern since such 13 loads represent approximately 20% of the forces and moments experienced 14 during LOCA. Such loads, even considering flexibility, can amount to a 15 significant portion of the loads reacted by the anchors. This problem 16 was not addressed by the SIT.
u 1 In reference to the LOCA, the procedures utilized by the Applicants 2 and approved by the SIT are spurious and therefore cannot be utilized '
3 to establish the justification for engineering judgement which concludes 4 that LOCA-induced loads are negligible. (See CASE Exhibit 761,4/26/83 5 Surrebuttal Testimony of Jack Doyle.) In addition, the friction which 6 exists between base plate and concrete must be considered in order to 7 arrive at an accurate stiffness value. Friction values of "0" are conser-8 vative for bolts but unconservative for the frame.during LOCA.
9 In order to detennine the k factor at a joint, many variables 10 need solutions, not the least of which is precisely how much friction 11 exists at the plane between plate and wall / slab interface.
12 Realistically for dynamic loading a friction joint is desirable.
13 See CASE Exhibit 763A, "FFTF Report, Drilled-In Expansion Bolts Under Static 14 and Alternating Load, BR-5853-C-4, January 1975." This is a report which I 15 obtained from Hilti Corporation regarding the use of Expansion Bolts 16 in the Fast Flux Test Facility in Richland, Washington. On page 15, item
! 17 1, bottom of page, it states: "A pair of spring washers should be placed 18 on every expansion bolt according to Figure 3. These spring washers 19 should be sized to produce a bolt tension of (1 10.4) S, where S is the i 20 maximum allowable anchor value." (The purpose of this suggestion is to
. 21 insure a friction joint to the allowable of the bolts in order to prevent 22 slippage.)
23 In answer to the statement on page 18 (SIT Report, end of first 24 paragraph) ". . . thermal loads are neglected when they are secondary 25 and self-limiting in nature and when the material is ductile," the loads I
r 1 experienced by the Richmond insert or Hilti bolt are not themal loads 2 but are loads induced by the resistance to the expansion of structures.
3 These forces are the result of displacement of the Richmond insert or the 4 Hilti bolt, and have nothing to do with themal conditions within the bolt 5 or anchor. This is also true for members which resist the expansion of 6 themal growth of framing members. Further, the NRC SIT (page 20, third 7 paragraph) concedes that ACI 349-80 requires the consideration of forces 8 caused by themal effects.
9 The current design of Hilti joints (bearing) is detrimental to 10 seismic analysis. To create a joint that is compatible with two totally 11 different loading conditions requires a balancing of philosophies. For 12 dynamic conditions, slip joints can be disasterous but for LOCA environ-13 ments slip joints are desirable. Therefore, the proper design for joints 14 under the two differing requirements is to produce a friction joint of 15 predictable capacity and accomodate thermal differentials by slip when 16 the designed capacity is exceeded. (See CASE Exhibit 763A.)
17 The Applicants go to great lengths to prove that the joints are 18 not of the friction type in writing off LOCA, but in writing off problems 19 related to seismic conditions, the joints become stiff with support frequencies 20 of 50 to over 100 Hz (deflections of .001 to .004 or less). See CASE Exhibit 21 763B, REF.: NRC Site Meeting - January 18, 1983, page 2 of 6, middle of l
22 second paragraph.
The tube Richmond philosophy is unpredictable under all conditions.
23 And according to the Applicants, the Hilti's are also unpredictable connections 24 25 (because they are also bearing joints).
t
1 On p' age 19, item (2) , the SIT discusses derivation of shear 2 allowable loads, but the actual shear loads are not discussed. If these 3 are bearing joints, why consider all bolts active in shear. This is 4 not proper engineering procedure.
5 On page 22 of the SIT Report, it is stated "Due to the high 517 tty 6 margins used for the design of anchors using Hilti-bolts, the resulting 7 small load from LOCA-induced thermal expansion would be unimportant."
8 The SIT's conclusion that the loads on Hilti's are unimportant does not 9 include valid calculations to back up the assumption. Our position is that 10 . the loads are significant (see especially CASE Exhibit 761C, page 8).
11 12 13 14 15 16 17 18 f 19 l
l 20 l
21 l
22 l
23 l
24 l
25 i
i 1 Q: Do you have cannents you wish to make regarding item d. beginning 2 on page 23 of the SIT Report?
3 Item d. Differential Thernal Expansion Effects in Pipe Supports 4 A: Yes. The argument that the stresses and loads resulting from 5 constraint of expansion are not significant is based on material contain-6 ing gross errors and therefore is invalid. Again the Hilti's and Richmonds 7 experience forces and moments induced by displacement and not internal 8 thermal considerations.
9 Also, the Applicants base their arguments on the flexibility of 10 the joints. If the spring factors are true for LOCA, they are also true 11 under normal operating conditions and SSE (Safe Shutdown Earthquake) con-12 di tions . If this is true, the generic stiffness used for pipe analysis 13 is no longer accurate and the loads resulting from such analysis are
! 14 without meaning.
15 16 17 18 19 20 21 22 23 .
24 25
1 Q: What comments do you have regarding item e. beginning on page 2 24 of the SIT Report?
3 Item e. Differential Thennal Expansion Effects in Wall-to-Wall, Floor-4 to-Ceiling, and Floor-to-Wall Pipe Supports 5 A: Again I refer to CASE Exhibit 761C. The Applicants' formulas 6 are inaccurate and incomplete. For the NRC to use such documentation to 7 preclude alternate means of determining the significance of an allegation 8 is a dangerous practice. Therefore, the conclusions of the SIT as ex-
,9 pressed in items (1), (2), and (3) of paragraph 2 on page 25 are without!
10 merit.
11 As to seismic differential displacement, this problem is agreed 12 to exist by the Applicants and the NRC Staff if the support goes from 13 floor-to-ceiling or wall-to-wall . However, they do not concede that a 14 problem exists for spans that go from wall-to-floor or wall-to-ceiling.
15 The differential displacement in reality could be as significant with the 16 wall-to-floor or wall-to-ceiling configuration, because if the floor 17 and ceiling or the floor and walls are not moving at the same rate or in 18 the same direction, then necessarily a differential will also exist between 19 the wall and floor or wall and ceiling. A few thousandths of an inch 20 displacement of one joint of a frame relative to oth'e r joints of the 21 same frame can have a significant impact on the forces and moments within 22 the frame.
23 The relative displacement between two slabs or two walls is 24 less severe than can exist between a wall and a slab. This results because 25 of the relative stiffness of the two. A slab is vertically flexible and
(..
I horizontally stiff. A wall is vertically stiff and horizontally flexible.
2 On page 26 (SIT Report, first full paragraph), the SIT states t
3 that "The inconsistency was identified in late 1981 in the normal process 4 of design review...In subsequent discussion, the Applicant showed the 5 Special Inspection Team component modification cards (CMC) 46174, Revision 6 8, and 46730, Revision 4 showing that the bottom portions of Item 25 on 7 support SW-1-132-701-Y33R (Doyle Deposition Attachment 7C) and Item 22 8 on support SW-1-132-703-Y33R (Doyle Deposition Attachmdnt 70) respectively,
,9 are to be cut off to eliminate the floor-to-ceiling columns on the east 10 end of each support."
11 If this is so, why were we doing a STRUDL analysis in very late 12 1981 and early 1982? How was this problem identified in 1981 (there is 13 apparently no NCR) when the CMC revision which identified the thermal problems 14 were written 12/82 and as stated above, I worked on one of these supports 15 in early 1982 and was still applying thermal until ordered not to use thermal.
16 CASE Exhibits 763C-1 through 763C-5, as I understand it, were .
17 supplied by the Applicants on disco.very to CASE; they are for CMC No. 46730 18 and all revisions, regarding support SW-1-132-703-Y33R. As shown on CASE 19 Exhibit 763C-4, it was not until Revision 3 was done on 12/29/82 that the 20 thermal problem was identified.
21 Similarly, CASE Exhibits 7630-1 through 763D-9 are for CMC No.
22 46174 and all revisions, regarding support SW-1-132-701-Y33R*. As shown 23 on CASE Exhibit 763D6, it was not until Revision 5 was done on 12/7/82 ,
{
- NOTE BY CASE: When we were preparing copies of these Exhibits, we discovered that we had not received Revision 4 of this CMC. We have re-quested it from Applicants but it will not be received and copied prior to mailing this testimony. We will send copies
.as soon as we receive Rev. 4 and can copy it.
i l that the therma'l problem was identified.
2 It should be noted that, even when the problem was identified, 3 what was done were only fixes; the problem was still not specifically
- 4 identified as being a thermal problem.
5 As far as creep is concerned, creep and . additional deflection 6 resulting from new live loads (or support reactions from the other side 7 of the slab) will be reacted by any support component tied to such slab.
8
! .9 10 11 12 13 14 15 16 17 18 19 j 20
'21 22 i
23 24 25
9_
1 Q: Do you have any comments regarding item f. beginning oc page 27 26 of the SIT Report?
3 Item f. Stability of Pipe Supports Designed for CPSES 4 A: Yes. A three-bar link is unstable regardless of what supports 5 exist up or down stream. My statement quoted by the NRC SIT (page 27, 6 last line through page 28, first four lines) was taken out of context.and 7 referred to a clamp strut situation and not three bar links. If the 8 SIT had read the entire article, they would have seen that my statements 9
were designed to show that three bar links are unstable under any circum-10 stances, because rotation of the frame surrounding the pipe will occur 11 without lateral displacement of the pipe.
12 If frictional forces are relied upon as a means of rotational 13 stability, data must be provided to prove that such friction is adequate 14 under all possible conditions. The effects of cyclic thermal reversals, 15 elasto plastic deformation, forces induced by self-weight acceleration of the 16 hardware, etc. , mujt all be included. Without such an analysis, all other l 17 arguments are gueiswork at best. In addition, the analysis must also 18 include the stress induced into the pipe due to the " clamping force."
l 19 Beyond this, in a later section of the SIT Report, the SIT states there is no thermal probeim because construction tolerances usually result in 20 a diametrical gap of up to 1/32 inch. (SIT Report, top of page 34.)
l 21 22 23 .
24 25 l
l l
+
1 Q: What comments do you have regarding item g beginning on page 29 2 of the SIT Report?
3 Item g. Use of U-Bolts in Pipe Support Design.
4 A: The first point made by the SIT involves the thennal and seismic 5 movement in the unrestrained direction for U-bolts used as one-way con-6 straints . The constraint of thermal expansion of up to 1/16 inch plus seismic 7 constraint of up to 1/32 inch is considered as negligible by the Applicants 8 and the SIT for all pipe sizes. A configuration such as this is shown in 9 CASE Exhibit 669B, items 3F and 3G, Mark No. CC-1-065-001-S33R, which is 10 an 18" pipe with a U-bolt. The stiffness of the pipe is such that 3/32 inch 11 displacement could result in very high constraint loads that can only be 12 determined by lengthy investigation and analysis based on the system, or 13 as determined by computer analysis restraining the point laterally. In 14 any event, the loads resulting laterally must be combined with the loads 15 normal to the direction of the U-bolt in an interaction equation supplied 16 by the manufacturer for the U-bolt in question. For this 18" U-bolt, 17 the allowable late'ral load is only a minor percentage of the normal load.
18 For PUS 180, the allowable lateral loads are for normal upset 320, Emergency 19 425, and faulted 500#. See CASE Exhibit 669B, item 13-0.
20 On the point made by the SIT on page 31 (last sentence) that the loads on U-bolts are within the allowables established by the manufacturer, 21 22 this argunent has nothing to do with displacements but only with stress and loads. The manufacturer is not interested in displacements and therefore 23 24 has no criteria for displacements. I ran a U-bolt for a 12-inch diameter 25 pipe through the STRUDL by the method of segmenting and at design loads
. 1 found the displacement to be in excess of 1/8 inch. I passed this informa-2 tion on to the people on site (at Comanche Peak), and I also had several 3 discussions with Gary Krishnan over this and other problems with U-bolts.
4 (See Regulatory Guide 1.124, CASE Exhibit 743 which was attached to CASE's 5 4/20/83 Brief Regarding Consideration of LOCA in Design Criteria for Pipe 6 Supports, page 1.124-3, i tem 6, Deformation Limits.)
7 U-bolts are not in compliance with IE Bulletin 79-02 (see NRC 8 Staff Exhibit 201, Supplement 2, Page 2 of 2, " Clearances") This is an 9 NRC mandate for radial clearance for pipes. See also CASE Exhibit 721, 10 page 1 of 15, " Clearances," Paragraph 2.2; this is the FSAR provision for 11 clearances.
12 The SIT neglects to treat the preloading of U-bolts with the 13 respect due. They dismiss the problem by stating that the torque amounts 14 to "the tightness attained by a few impacts on an impact wrench or the full 15 effort of a man using an ordinary spud wrench." (SIT Report, page 32, 16 fourth paragraph.) Assume the length of a spud wrench to be 10" to the 17 center of pressure exerted by this man. Assume further that this man 18 exerts about 80 lbs. of force. The stress in a 3/4" diameter (rod size) 19 U-bolt would be about equal to 80 times 10 divided by half the thread diameter 20 and this amount divided by the tangent of the sum of the angle of friction 21 and the slope of the thread. For the U-bolt in question, the stress in 22 the U-bolt would be about (at its tensile area of .334) 2580 divided by 23 tan.14 degrees (friction angle) plus 3 slope of thread ((1/thds per inch) .
24 / circumference of root diameter). This equals 2580 divided by .3046 equals 25 force equals 8472 pounds. Therefore, stress equals 8472 divided by tensile
)
area (.334) equals 25366 PSI. How can this be written off when this 2 approach is only a rough cut and not a precise evaluation of the problem --
3 it indicates very high stresses and shows a potentially severe problem is 4 present. As a matter of fact, the ITT Grinnel Company has one U-bolt type 5 (B. metallic to overcome expansion (thennal) problems) of clamp which contains 6 a caveat for over-torquing. See CASE Exhibit 763E.
7 A second factor involves the use of SA307 bolt material for friction 8 joints. This material is not allowed for friction joints. See CASE Exhibit 9 752 ( ASME Appendix XVII, Table XVII-2461.1-1, page 388; this was attached 10 to CASE's 4/20/83 Brief Regarding Consideration of LOCA in Design Criteria 11 for Pipe Supports), where NOTE 1 states " Friction type connections loaded 12 in shear are not pennitted. The amount of clamping force developed by 13 SA-307 bolts is unpredictable and generally insufficient to prevent complete 14 slippage." Beyond this, A307 is not recommended for dynamic applications, 15 as may be noted in Salmon and Johnson " Steel Structures Design and Behavior,"
16 Chapter 4 Structural Fasteners, at 4.1 Types of Fasteners. At the bottom of 17 page 85 and continuing on page 6 (in reference to A307 unfinished bolts), these 18 bolts are recommended for " loads (which) are primarily small and static in 19 nature" (by inference not recomended for dynamic loads). See CASE Exhibit 20 763F.
21 Additionally, the inspection procedures used by the Applicants 22 to insure that no cracking of the U-bolt occurs during cinching, while 23 comendable, does not explain how an inspection can be performed when this 24 U-bolt is subjected to thermal expansion of the pipe and the design loads.
25 In the last paragraph of page 32, the NRC states "In the case
1 of insulated piping (e.g., main steam, feedwater, and residual heat removal 2 piping), the temperature differences between the U-bolt and the pipe will be 3 negligible . . ." This is nonsense. In my deposition / testimony (CASE Exhibit 4 669B, items 13E through 13J), I submitted test data which (although the tem-5 perature range is 9000 F, the temperature is a constant in the equation, with 6 a minor variation in the coefficient of thennal expansion, or CTE) indicates 0
7 a minimum of 15% variation overall between the Figure 244 clamp (360 contact) 8 and the pipe. At PT 2 (direct contact point of clamp) outside diameter, 9 differential equals 3.5%, at PT. 4 (tangency PT. of pipe and clamp) approximate 10 differential is as little as 10% using this value. This would still mean the 11 differential thennal expansion is about 20 thousandths of an inch (and the 12 resistance factor for 900c insulation is better than 6000 insulation).
13 The drop in temperature for metal in direct contact with the pipe and 14 insulation may best be noted in CASE Exhibit 669B, item 13H. Four inches 15 from the pipe, the temperature has dropped 25%. The U-bolt on the main 16 steam loses contact (optimum) at the horizontal tan. pt.; therefore, at a 17 minimum,16 inches of the U-bolt is not in contact with the pipe. What are 18 the stresses in the pipe and U-bolt which resist this force? In the NRC Staff's 19 rebuttal testimony in the September 1982 hearings (NRC Staff Exhibit'201), Mr.
20 Tapia stated, in Answer 10 on page 6, that effects induced in the pipe wall by 21 a support must be included with the effects derived from pipe stress run. The 22 question as to what the stresses in the pipe and U-bolt are which resist the 23 force was not answered by the SIT, and if the NRC ever answers the question they 24 must keep in mind that these stresses are additive with the pre-tension stresses 25 caused by torquing and all of this is additive with the design load stresses.
It is because of these thennal expansion problems that the Grinnel Figure 26 iWc=
1 224 would work; stainless steel (the material of their U-bolt) has a higher 2 coefficient of thennal expansion (CTE) and therefore will not act as a 1
3 constraint to pipe growth.
4 The same problems noted for the U-bolts on the main steam exist 5 for all the cinched-up U-bolts because, although the temperatures are 6 lower, the lack of insulation means a higher temperature differential between 7 the clamp and the U-bolt.
8 Besides other problems, the U-bolt has two additional features 9 which offer problems: (1) Local deflection due to the weak curved BM which 10 is not in 1800 contact with the pipe; and (2) local stress in the pipe due 11 to point contact. This could result in plastic deformation.
12 The stress levels in the pressure boundary as a result could be 13 excessive due to point loading. See CASE Exhibit 669, p. ages 195/5-25,196/1-21, 14 197/13-19, and 319/1-11. This must be considered in order to comply with 15 the provisions of ASME Code sections NB-3645, NC-3645, and NF-3127; see also 16 NB-3613.3 and NC-3600.
17 In reference to the box frames (last paragraph, page 33 of SIT 18 Report), the SIT noted that the expansion in four directions would be about 19
.008 inch and stated that stresses due to such expansion would be negligible.
(
20 The box structure around the pipe is a continuous frame. Any load on one 21 member creates fixed-end moments which relative to the stiffness characteristics 22 are distributed around the frame. However, when loads of equal magnitude 23 are placed at similar points on the four members, the distribution balances 24 and each member acts as a fixed ended beam. To deflect a TS 6 x 4 x 1/2 25 tube 24 inches long with fixed ends (see CASE Exhibit 6698, items 41, 4J,
- 1 4K, and 4L) .004 (considering that the pipe takes the other .004) takes 2 a considerable equivalent force; for example, .004(192)(278900000)(17.6)/24 3 3 equals 27,280 lbs. See AISC 2-203 (we have deflection solved for "P"). See 4 CASE Exhibit 763G. Is this 14-ton force insignificant? (Obviously,because S of stiffness differentials the pipe will see most of the actual force, but 6 this indicates the magnitude of the problem.
7 In reference to the inclusion of loads on piping, see the NRC 8 Staff rebuttal testimony in the September 1982 hearings (NRC Staff Exhibit 9 201, Answer to Question 10 on page 6), in which Mr. Tapia references the I 10 ASME Code sections that require consideration of induced loading: NB-3645; 11 NC-3645; and NF-3127.
12 Finally, in the first paragraph on page 34 (SIT Report), the SIT 13 states that thermal expansion would be less likely to be a problem since 14 "although the design may specify a zero clearance (no gap) for a box frame, 15 construction techniques often .(but not always) result in a diametrical gap 16 of up to 1/32 inch." However, the SIT relies on "0" gap and frictional 17 forces to prevent clamp rotation (included among this problem are box frames);
18 see page 28, second full paragraph of SIT Report.
19 20 21 22 23 24 25
~
. 1 Q: Do you have coments regarding item h., beginning on page 34 of 2 the SIT Report?
3 Item h. Loading Due to the Seismic Acceleration of the Pipe Support 4 Structure 5 A: Yes. The SIT failed to note that the effects of seismic accelera-6 tion are additive to the nonnal design loads. Further, the ASME Code, 7 Section NF-3111(e), requires the inclusion of seismic effects on the frame 8 (see CASE Exhibit 659B, attachment to Mark Walsh's testimony).
9 In the NRC Staff's rebuttal testimony in the September hearings 10 (NRC Staff Exhibit 201), Mr. Tapia stated that PSE guidelines called for 11 seismic consideration of the support and hardware; see page 7, Answer 13, 12 continued on page 8, which states 'that a future program will require accelera-13 tion in all three directions. In Answer 14, page 8, Mr. Chen agrees with 14 Mr. Tapia that a seismic analysis will be conducted in the future. Beyond 15 this, the ASME Code, Sub-Section NF-3112.2(b) mandates this; also NF-3111(e) 16 and Criterion II of 10 CFR Part 50, Appendix A. Applicants' FSAR Section 17 3.7B.3.5 (Applicants' Exhibit 3) states that lacking knowledge of the funda-18 mental frequency, the peak of the appropriate response spectra must be used.
19 At item (3) on page 37 of the SIT Report, the SIT indicates that 20 130 and 100 pound loads were neglected in the analysis of pipe systems.
21 The accelerated effect of a 130 pound load would have an effect on the up 22 and downstream supports as well as the overall output for the pipe stress.
23 IE Bulletins 79-14 and 79-02 both address the accuracy of pipe l
24 loads. See NRC Staff Exhibit 201C for citation in reference to IE Bulletin 25 79-04 valve weights. In the case of SI-1-104-008-C53K, the 100# before i
1 acceleration represents about 3% of the total loading. In the case of 2 SI-120-004-C52K, the 130 lb. unaccelerated lo;d represents about 4% of the 3 load. (SIT Report, page 37, item (3).) Are we to believe that possibly i 4 10% of the load on pipes and supports can be routinely neglected? Obviously 5 it cannot.
6 -
7
- 8 9
10 11
- 12 13 14 15 16 17 18 19 20 1
21 22 23 24 25 1
. 1 Q: What comments do you have regarding item i. beginning on page 38 2 -of the SIT Report?
^
3 Item i. Moment Restraints and Local Pipe Stress Due to Welded Stanchions 4 on Pipes 5 A: How can we conduct a rebuttal to the Applicants or to the NRC 6 if each time we point out an error, either one or the other evades the 7 issue by stating things like "this rotational restraint had also been iden-8 tified during the Applicants' normal design review"? The fact remains that 9 the issue was .?ised and was a proper allegation at the time. The fact 10 that subsequent tt the raising of the issue, the Applicants state "oh, we 11 know that" does n(t alter the fact that as of July 1,1982, this problem 12 was incorporated in;9 the systems at Comanche Peak. And .the major question 13 of why remains unanswered.
14 ,
In addition, the SIT Report states that the Applicants know the 15 problem exists but have only addressed it by overlapping assumptions.
16 The pipe stress analysis for the main steam with a moment restraint as 17 indicated has not been rerun (SIT Report, page 39, first paragraph, second 18 sentence).
19 20
,21 22 23 24 25
_ 19 _
1 Q: Do you have comments regarding item j. beginning on page 40 of 2 the SIT Report?
3 Item J. Deflections and Local Stresses in Pipe Support Structures 4 A: Yes. The SIT (Page 41, last paragraph of SIT Report) discusses 5 support CC-1-107-008-E23R and claims that tests have shown that the actual 6 deflection is less than 1/16 inch.
7 The Applicants have submitted calculations (finite element) which 8 resulted in .0608 inch deflection (see CASE Exhibit 763H). The calculations 9 indicate that tests show deflections of .055 inch (by exactly which methods 10 and procedures we are not sure). In any event, the load used was 153 lbs.;
11 the load on the support at the time of the last hearings was 209 lbs. (see 12 CASE Exhibit 669B, item llTT). Even using 73% of the original load, not 13 including the accelerated hardware and structural steel weights and using 14 the most sophisticated method (and least conservative methods), the support 15 barely makes it. Including the acceleration of hardware and steel as re-16 quired may prove the support inadequate even at 153# loads. Additionally, 17 the stiffness factor for this support is about 2500#/in. -- hardly the generic 18 stiffness used in pipe stress.
19 Beyond this, the drawing we have shows a negative load of 153**
20 (used in test and calculations) but it also shows a positive load of 1177 21 lbs. If the larger load were used, the deflection would be almost 1/2 22 inch. And it is the larger of the two loads which must be used.
J
- NOTE FROM CASE: The copy which we received of the drawing was very poor quali ty. We have requested a better copy from the Appli-cants, but it will not be received in time to send with -
this testimony. We will send a copy of the improved version as soon as we get it from the Applicants.
1 The SIT agress (SIT Report, page 41, paragraph 4) that an over-2 deflection exists for the diaphram on support CC-2-008-709-A43K, but further 3 states that there are no overstressed conditions in the weld. The moment
-4 developed in the plate by the deflection in the plate starts at this weld, 5 and I find no calculations that indicate the effect of Horizontal shear 6 flow on the weld. I can only assume that the SIT conclusions are by engi-7 neering judgement. At least the NRC found the problem was correctly stated 8 as far as stress and deflection are concerned.
9 In reference to stress in the W6 x 12 beam (SIT Report, page 10 42, paragraph 2), the analysis used an optimistic approach assuming the 11 whole length of the web was effective; and stated that a stress of 184 12 KSI was present in the web.
13 Although the SIT appears to have missed a second similar problem, 14 I stand corrected as the result of the SIT's correctly pointing out that 15 I used the cube of the web thickness and not the square in my calculation.
16 However, the stress in the web for the first problem of CC-1-028-034-S33R 17 under optimistic conditions and exclusive of the compressive stresses re-18 mains at a minimum of 184 (.23) = 42.5 KSI. This is still an overstressed 19 condition for nonnal operation (.75 Fy equals 27 KSI). In addition, 20 continuing deflection will increase the component load seen by the minor 21 axis bending.
l l
22 The second similar problem not addressed by the SIT (see CASE Exhibit 669B, item 11QQ) was regarding CT-1-017-034-S35R. I assumed that 23 due to vertical seismic and thennal loads, the node could displace 1".
24 (There is no documentation to back this up, but there will be some displacement.)
25 26 Again using the full web length, the load (using the proper section modulus) 27 would result in 119.32(.23) or 27.5 KSI per inch of node displacement.
1 As little as one quarter inch will result in stresses of at least 7 KSI 2 which has.not been included in the calculations or considered by the SIT.
3 These two members should have stiffeners at a minimum and should be addressed
.4 by the NRC.
5 In these two cases of transferring a load f. rom the pipe to the 6 wall, the reacting member is not the W6 x 12 as much as it is a plate on 7 edge which is not a very stable member for the indicated forces of over one 8 ton, especially if there is any loading across the plate.
- 9 In referring to these two problems, the SIT found an error in my 10 unchecked rough calculations and without further checking dismissed the 11 allegation.
12 Paragraph 3 (page 42 of SIT Report) refers to CS-239-007-A42R, 13 which had a plate that was overdeflecting, and on June ll,1982, a CMC i le (component modification card) was issued to increase the plate thickness 15 (according to the SIT Report). The fact remains that a't the time of my 16 allegation the problem existed, and was generic (to Comanche Peak) in 17 nature, and had a valid technical basis.
l 18
! 19 20 21 22 ,
23 1
24 25
1 Q: Do y,ou have any coments regarding item 1. beginning on page 44 2 of the SIT Report?
3 Item 1. Consideration of Kick-Loads 4 A: Yes. On page 45, paragraph 1, the SIT states that I am incorrect 5 in assuming that kick-loads are not considered in the analysis of piping.
6 At the time I left Comanche Peak, the kicks were not included (see CASE 7 Exhibit 669B, items llRR and llSS, Support CC-1-057-006-A33R). This support 8 has a 12030' angle to the vertical. The load shown on the drawing is in 9 the Fy direction only. If the skew position of this support were incorporated 10 into the computer analysis, the loads would have loads Fy and Fz. From CASE 11 Exhibit 669B, items 13T and 130, the drawing shows an angle of~ 39 0 with a 12 load on the pipe of Fx 2456 nomal upset. If there is no kick, where is 13 the load Fy? (See, also CASE Exhibit 669B, items 13FFF and 13GGG.)
14 Beyond this, the configurations shown are unstable. Any variation 15 in the Fy or Fz loading or in fact the self-weight excitation of the clamp 16 and hardware will induce instability. The degree or importance of this insta-17 bility is less important than the fact that for nomal upset conditions a 18 support configuration of. this type was selected rather than a support con-19 figuration which offers stability without external assistance; i.e. , the 20 pipe. Under several loading conditions, these supports are effectively l 21 non-existant; for example, referring to the drawing (CASE Exhibit 669B, 22 Items llRR and 115S), if the seismic force components are vertically down 23 and to the east (the force vertically down also is additive to the gravity 24 load).
25 The SIT offers " proof" that this support has no problem (see CASE Exhibit
1 763B, NRC Site Meeting notes dated 1/18/83). Two ADLPIPE runs were checked 2 for skew cards and were found to be incorporated into runs dated November 5,
, 3 1982, and September 10, 1982, long after I brought up my allegations.
I 4 i 5 7
8 l 9 10 11
! 12 13 14 l
15 .
16 17 18 19 20 21 22 23 24 25 l
1 Q: What coments do you have regarding item p, beginning on page 49 2 of the SIT Report?
3 Item p. Welded Stepped Connections, Fillet Welds and Skewed Welds 4 A: In reference to the SIT's statement that welding is controlled 5 by ASME, not AWS, the argisnent is less than accurate. The Applicants and 6 the SIT are adamant in stating that AWS is not a requirement at CPSES, and 7 insist that Secion IX of ASME controls. (See Applicants' Exhibit 142F, 8 page 3, Mr. Finneran's Answer 8; also page 7, Answer 22. Further, see 9 tr. 5074/4-13.) The PSE Guidelines see the situation somewhat differently, 10 however. See CASE Exhibit 716,Section XI, Weld Calculations, of PSE 11 Guidelines, dated 5/11/82 and signed by Applicants' witness Finneran.
12 On page 1, paragraph
2.0 REFERENCES
, D., American Welding Society Code 13 D1.1 is cited. As for Section IX of ASME, it does not address the analysis 14 of welding. The provisions of Section IX address: Article 1, General Require-15 ments: weld positions, mechanical tests, acceptance criteria, and examination; 16 Article 2, procedure qualifications; Article 3, performance qualifications; i 17 Article 4, welding data including welding variables, joints, base metals, 18 filler metals, etc.
19 The treatment of welding is similar to the treatment of structural l
20 analysis of pipe supports. That is (in the case of structural analysis),
21 the ASME Code, Section NF and Appendix XVII, contain many procedures for
~ 22 the analysis of structural steel which were adapted from the AISC Manual.
23 I'n the event a particular procedure or practice is not contained with the 24 ASME material, one must refer to the source, i.e., the AISC Manual, for 25 answers to such problems. This is also true for welding with this caveat:
l
I there has been less material incorporated into ASME from the AWS source than 2 has been incorporated from AISC.
3 The mere fact that material has not been incorporated from a 4 particular source into the ASME Code does not preclude the validity of 5 such source material, nor does it follow that it may be excluded from 6 consideration.
7 Up until the time I left Comanche Peak, many procedures in reference 8 to welding were ignored; for example, minimum weld, reduction of effective 9 throat, Beta (ratio of diameters or sizes of a square tube) factors, among 10 others . The fact that subsequent to my leaving Comanche Peak a procedure 11 has been introduced does not alter the fact that for many years these pro-12 cedures were ignored although they existed in the source documents.
13 The SIT states (page 50, first paragraph under " Fillet Welds" 14 of SIT Report) that MS-028-034-S63R (CASE Exhibit 669B, items 13W and 13X) 15 was evaluted for minimum weld size. On the drawing, the engineer calls 16 for 1/4 inch fillet all around with a plate thickness of 1-1/2" and a tube 17 wall thickness of 3/8 inch. The minimum weld required by AISC, AWS or 18 ASME is 5/16 inch. The same is true of SI-031-704-A32R (CASE Exhibit 669B, 1 19 items 12H and 12I) . The weld to base plate here is also 1/4 inch fillet l
20 all around, the plate is 1 inch thick, and the tube wall is 1/2 inch; 21 therefore, the minimum weld required is 5/16 inch. Both of these drawings 22 are marked "As Built." If this weld is now 5/16", the SIT must offer docu-23 mentation to prove this and the procedure followed to achieve the weld.
24 In reference to the width ratios (page 50, paragraph just above
" Fillet Welds" of SIT Report), the SIT _ states that the lowest width ratio I
25 l
_ L_ _
4 1 used at Comanche Peak is 0.67 and that therefore no joint rotation or punching shear problems are present. There are two fallacies in this 2
3 argument. First, the width ratio in common use until the report of r
4 -s September 1982 was Blodgets Ratio of .8. Second, the NPSI rear brackets 5 which are mounted on many tubes (for example: CC-1-il6-038-F43R, CASE 6 Exhibit 669B, items llWW and 11XX) present a ratio of less than .4, and therefore require a full analysis for punching, rotation and deflection.
I 7 8 Further, supports MS-1-146-700 and MS-1-147-700 have TS 6 x 3 tubes framed 9 into TS 8 x 8 with the 6" dimension along the axis of the 8 inch tube, 10 "W" factor .375 not .67. This is only to mention two supports.
11 12 13 14 15 16 17 18 i
l 19 20 21 22 23 24 25 l
\
l Q: Do you have comments you'd like to make regarding item q, beginning 2 on page 52 of the SIT Report?
3 Item q. Section Property Values Utilized by PSE 4 A: Yes. The SIT compared the values of the Cold Form Institute steel 5 to the 8th Edition AISC steel and found differences of 7.1 percent difference 6 in stress for a 4 x 4 tube and 7.5 percent difference for stiffness.
7 Without considering the difference between AISC 7th Edition steel (of which 8 large quantities were used at Comanche Peak) and the Cold Form Institute 9 steel (see CASE Exhibit 763I) which indicates a 20% differential in stiffness 10 and 11.4% difference in section modulus, does the NRC expect us to believe 11 that even an 8% differential in stress and stiffness is acceptable for 12 engineering nuclear supports? Obviously this cannot be assumed to be correct.
13 These differentials are starting to add up: 8% for steel properties,10%
14 for masses on the pipe, neglect of seismic for hardware and support steel, 15 seismic differential displacements, creep and live load displacement of 16 concrete, thermal effects, etc. Individually these " oversights" may not 17 appear to be significant to the uninitiated; however, even the "somewhat i
18 knowledgeable" must recognize that collectively the problem is significant.
19 (See cross-examination t.estimony of Applicants' witness Finneran, tr.
20 4953-4985, especially 4962/23-25.)
21 22 23 .
24 25 i
r
I Q: What cannents do you have regarding item r, beginning on page 53 2 of the SIT Report?
3 Item r. Support Pads Welded Over Pipe Girth Welds 4 A: At the time I received the sketch and math model to run on STRUDL, 5 the valve was a Class 2 item (see CASE Exhibit 669B, items 2E and 2F),
6 regardless of what the classification is at the present. Beyond this, 7 when I returned the support for consideration of ASME Section XI, that 8 was the last I saw of it. Gibbs and Hill document 2323-HS 46A, pages 3-44 9 and 3-45 are more stringent than Appendix XI. The fact that it is now in 10 a different configuration does not alter the fact that one of the engineers 11 designed a support (Class 2 at that time) which was not in conformance 12 with the ASME Code and additionally induced moments into the Class 2 weld.
13 I only mentioned-this support briefly in my deposition / testimony (CASE Exhibit 14 669) to indicate that there were people working on site who were less than 15 "somewhat knowledgeable."
16 17 18 19 20 21 22 23 24 25
1 Q: Do you have connents regarding item s, beginning on page 53 of 2 the SIT Report?
3 Item s. Damage to Pipe Support During Hydrostatic Testing 4 A: Yes. The SIT goes to great length by deductive reasoning to pre-5 clude all possibilities of damage to support CC-1-ll6-038-F43R other than 6 the reason given by the Applicants. The rationale behind the SIT's reasoning 7 defies logic. Their claim is that an NCR (nonconformance report) was written 8 in October of 1980 because of a bent eye rod on the strut and they assume 9 that, although the bent tube wasn't included in the NCR, it must have 10 happened at the same time. This story has a few flaws, the principal one 11 being rather critical if true. The NCR process must be woefully inadequate 12 if damage occurs and as a result of the NCR's only part of the problem 13 is corrected (or even identified). Additionally, why wasn't the unrepaired 14 damage reported in the ensuing three-year period?
15 In addition, regardless of how the damage occurred, the wall 16 of this 8 x 8 x 1/4 tube was expected to react a 4900 pound load for normal 17 upset conditions. The width ratio of the connection was well below the l 18 new magic number of .4 and deflection (locally) would make this a doubtful 19 candidate as a nuclear support. This damaged support is an indication 20 of the results that occur when the width ratio falls below .4.
1 l 21 22 23 24 25
1 Q: What comments do you have regarding item 4., beginning on page 54 2 of the SIT Report?
3 Item 4. Inspection of Vendor Certified Supports 4 A: The Applicants and the SIT would have us believe that the drawings 5 questioned at these hearings are only preliminary drawings and sketches.
6 The only sketch discussed in my deposition / testimony is the one at CASE 7 Exhibit 669B, items 2E and 2F.
8 As to the contention that these are only preliminary drawings, 9 if this is so, then why has the support on this drawing been fabricated, 10 erected, inspected, painted, insulated and (as in a number of other cases) 11 turned over and stamped "As Built"? Surely this is not in a desperate effort 12 to support the pipes. This is done by the crafts with temporary supports.
13 I believe this word " preliminary" is another in a group of words 14 ("i terative," " final as-built review," " design reverification," " thermal 15 stresses," etc.) designed to evade the point -- was it wrong at the time 16 of the allegation and if so, why? Also, if it was wrong, to what extent 17 are these kinds of errors incorporated into other systems?
18 19 20 l 21 l
l 22 23 24 25
1 Q: Are there any other concerns you would like to address?
2 Problems raised by Walsh/Doyle which were not addressed by the SIT 3 Q: Yes. Among the problems raised which were not addressed by the 4 SIT Investigation Report 50-445/82-26, 50-446/82-14 are the following:
5 1. From CASE Exhibit 6698, items 4Q and 4R:
6 A 2-3/4 inch Q U-bolt with 1/8 inch gap having an upload of l 7 101,456 pounds (nomal upset) through a double pin ended strut. The U-bolt 8 rotational capability insures collapse. Beyond this, the vertical load g down is 116,562 pounds (emergency). Determining the result of this point 10 load on the pipe wall, and the deflection of the U-bolt are required pro-11 cedure.
12 2. From CASE Exhibit 669B, items 4S and 4T:
13 Same problem as item 1. above except U-bolt is loaded with 14 an upload of 7466 and vertical down load is 35,620 pounds. This support 15 is also tied between a floor and wall, so seismic differential and themal 16 considerations are required as well as out-of-plane seismic effects.
17 3. From CASE Exhibit 669B, items 5B and SC:
1 18 The frame in this drawing is supported by pins so it is free jg to swing in the Z (East-West) direction. However, there are forces in three l 20 directions, two caused by themal movements; therefore the support is un-l 21 stable and should have been addressed by the SIT.
l 22 4. Fr m CASE Exhibit 669B, items 8E and 8F:
- 23 These are not supports but relate to supports. The Applicants ,
l l
24 and the SIT argue that the anchor bolt base plate system is not of the friction type. If this is so, the total bolt system cannot share equally 25 l
1
. 1 in shear and a,s shown on 8E as few as two bolts can carry the total shear 2 (really one can take all the shear and another take a second shear from 3 coupling out a moment). The Comanche Peak procedure divided (by computer) 4 the load (shear) to every bolt in the pattern. This problem also must be 5 addressed.
6 5. From CASE Exhibit 669B, items 8T and 8U:
7 This Class 1 support, which is displacement-critical, transfers 8 its load of over 1,000 pounds through a moment ann of 6 inches coupled g through 24 inches. We therefore have about 560 pound couple and direct 10 load (SRSS) in shear shifting the support in its oversized bolt holes 11 which are not acting as a friction joint. This problem must be addressed.
12 Slip joints, if used, must be considered in dynamic analysis as detrimental 13 to the prediction of results at best.
14 6. From CASE Exhibit 669B, itesm 8V and 8W:
15 This support cantilevered off an NPSI tube arrangement 16 can shift in its oversize holes, thus leading to excessive deflection.
17 This problem must be addressed. See Item 5 above. Also, an error in the 18 NPSI procedure to determine tension from Mx remains unanswered. See 19 CASE Exhibit 669B, item' 8P (center of compression). The method of analysis 20 f r coupling out the torsional moment in the tube (see CASE Exhibit 669B, 21 items 8-0 and 8P and CASE Exhibit 669,pages 123/10-25 and 126/1-4.) The center 22 f pressure for a 4 x 4 tube is not 2(2/3) but rather is 2/3(2-2t). This 23 is because of the radius which does not contact the washer. The couple 24 take place at the washer and the tangency point of the tube, not the washer 25 and concrete. See CASE Exhibit 669, page 125/1-25 and 126/1-9.
1 7. From CASE Exhibit 669B, items 10C,10D,10E,10F, and 10G:
2 The amount of friction which actually exists in the joints 3 using Hilti's must be known since any attempt at detennining stiffness 4 factors for LOCA relief depends on breaking the friction plane before dis-5 placement occurs. This is true of Richmonds to a much lesser extent, since 6 they are basically a slip joint. Therefore, the precise tightening (torque) 7 on the nuts must be available. This critical question must be addressed.
8 8. From CASE Exhibit 669B, items llJJ,11KK, and 11LL:
9 The instability of this support and the vertical kick induced 10 into the riser has not been addressed. What happens to the components
~
11 from the N.S. snubber offset of 9.80 and the E.W. snubber offset of 160?
12 9. From CASE Exhibit 6698, item 12N:
13 The themal constraint caused by the locked-in main steam 14 pipe will result in high moment loads on the structure and pipe stresses 15 of unknown magnitude (without tests to determine the precise temperature 16 gradients from point to point, analyses which tend to avoid conservatism 17 are less than adequate). This problem was not addressed, especially in 18 reference to the displacement loads induced into the structure and its welds, 19 i 20
- 10. From CASE Exhibit 669B, items 13YY,13ZZ, and 13AAA:
l 21 The problem of stiffeners reacting loads at the center 22 of tube walls, thus inducing bending and high stresses, has not been 23 addressed.
24
- 11. From CASE Exhibit 669B, items 14E,14F,14G,14H,141,14J, 25 14K,14L, and 14M:
i i
W 1 These anchors all rigidly lock the pipe and prevent thennal 2 expa nsion. Beyond this, the supports are tied generally from wall to slab.
3 They are therefore subject to differential displacements. The stresses
' 4 induced by thermal constraint must be considered in the analysis, as well 5 as differential displacement. This point has not been addressed by the 6 NRC.
t 7 8
9 10 11 12 13 14 15 16 17 18 19 20 l 21 22 23 24
, 25
F
. 1 Q: Mr. Doyle, do you recall whether or not there was anyone from the 2 NRC Staff present during your August 19-20, 1982, deposition?
3 A: Yes. Mr. Tapia arrived the second day just before I completed 4 the deposition.
5 Q: Did the NRC ask you any questions during your deposition?
6 A: No.
7 Q: Were you contacted by anyone with the NRC regarding your concerns 8 following your depositio'n and prior to the September 1982 hearings?
9 A: No.
10 Q: Were you contacted by anyone from the NRC regarding your concerns 11 following the September 1982 hearings?
12 A: No.
13 Q: So you were not contacted by anyone from the NRC Special Inspection 14 Team (SIT) regarding your concerns during the SIT's investigation of what 15 have come to be known as the Walsh/Doyle allegations?
16 A: No.
17 Q: In your testimony, did you at any time identify your concerns as 18 they are expressed in the SIT Report (i.e., as identified by the SIT as 19 items a through s)?
20 A: No.
j 21 Q: So the manner in which your concerns are identified and grouped 22 in the SIT Report was detemined by the SIT without any consultation with 23 you, is that correct?
24 A: Yes.
25
, 1 Q: Before we leave the subject of LOCA, Mr. Doyle, there a just a 2 couple of other questions which we have.
3 Is it your understanding that 10 CFR Part 50, Appendix A, NRC 4 Regulatory Guide 1.124, other NRC regulations, ASME and other industry 5 codes require that the effects of constraint of free-end displacement 6 (or thermal expansion) resulting from a LOCA be considered in the design 7 criteria for pipe supports?
8 A: Yes.
9 Q: To your knowledge, is there,any NRC regulation, ASME code, or 10 other industry code which states that these LOCA effects are not required 11 to be considered in the design criteria for pipe supports?
12 A: No.
i 13 14
, 15 16 17 18 19 20 1
i 22 23 1 24 25 1
e 1 Q: In your testimony, Mr. Doyle, you did not address the following 2 items which were identified in the SIT Report:
3 a. The Interfacing Between Pipe Support Design Groups; 4 b. Interfacing Between Pipe Support Design Groups and Pipe Stress 5 Analysis Organizations; 6 k. Consideration of Friction Loads; 7 m. Modeling of Wide Flange Members As Infinitely Rigid In Torsion; 8 n. Effect of Cold-Forming on the Ductility of Tube Steel; and 9 o. Operating Condition Loads Appear to Be In Error.
10 Does this mean that you agree with what is stated in the SIT 11 Report regarding those items?
12 A: No, I do not necessarily agree. It is a matter of priorities.
13 The items mentioned are of less consequence than the matters covered 14 -- in reporting a fire, do you explain to the Fire Department the nature, 15 extent, location and potential of the fire or do you get all excited about 16 a kitchen window which broke because of the heat?
17 Q: Do you agree with the Summary and Conclusions contained on pages 18 6 and 7 of the SIT Report?
19 A: Absolutely not.
20 21 Q: What is your opinion of Applicants' iterative design process?
22 A: The " iterative design process" is a phrase developed to explain 23 away the gross errors in engineering that have occurred over the years. -
24 The phrase implies that regardless of the stupidity involved; i .e., instability, 25 excessive deflections, overstressing (U-bolts, for example), induced loadings,
e .
1 neglect of required parameters, incomplete and erroneous formulas, etc.
2 -- it is of no consequence until the final as-built, verified and reveri-3 fied, vendor certified drawing is stamped, approved, reviewed, and blessed 4 by the pontiff. In short, gross error is proper engineering procedure for 5 nuclear power plants until we say the plant is finally, completely, totally, 6 absolutely, irretrevibly done, " Amen."
7 Consider these preliminary designs and sketches referred to in the 8 SIT Report (page 6, Summary and Conclusions). The second sentence reads 9 "They supported their allegations with a number of preliminary designs 10 and sketches for various supports." These preliminary designs were initiated, 11 checked, approved, issued to the field, fabricated, constructed, QC inspected, 12 bought off, revised as many as 12 to 15 times, painted, and the pipes insulated.
13 This doesn't sound very much like the effects resulting from a " preliminary" 14 sketch. If everything is so preliminary, why are the Applicants filing for 15 an operating license to eventually maintain over 15 billion ci (curies) 16 of radioactive material in each reactor plus hundreds of millions of curies 17 (mainly in transuranic products) in the spent fuel pool -- when no one knows 18 what the plant will look like when it's finally operating?
19 20 21 22 Q: Have you discussed all of the problems at Comanche Peak with 23 which you are concerned?
24 A: More or less, but then I am only acquainted with a minor fraction 1
25 of the total engineering and construction both off and on site.
e 1 Q: Are you aware of other problems at Comanche Peak?
2 A: Yes. However, my infomation is all hearsay and for that reason 3 I did not include it in my deposition / testimony of August 1982.
4 5
6 7
8 9
10 4
11 12 13 14 15 16 17 18 19 20 21 1 22 23 24 25
_ . . . _ . . - _ . - _ - _ _ _ - . - - - -