ML20083K067
| ML20083K067 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 03/09/1984 |
| From: | Williams N CYGNA ENERGY SERVICES |
| To: | Ellis J Citizens Association for Sound Energy |
| References | |
| 84042.04, NUDOCS 8404160053 | |
| Download: ML20083K067 (18) | |
Text
.
9
=.
101 Cah'ornia Street. Suite 100k San Francisco. CA 941115894 415'397-5600 March 9,1984 84042.04 Mrs. Juanitc Ellis, President gb-Citizens Association for Sound Energy I426 Sou;h Polk Dallas, Texas 7S224
Subject:
Comanche Peak Steam Electric Station independent Assessment Program -
Response to CASE Questions
Reference:
(1) Brief Summery of Generic Probleins from CASE Witness Jack Doyle, 2/22/84.
(2) Brief Summary of Cross-examination Questions from CASE Witness Mark Walsh,2/22/84.
Dear Mrs. Ellis:
Enclosed piecse find our cesponses to reference (1) items S,6,9,14 and 18; and reference (2) item I.
Further responses will be forthcoming.
Very truly yours, hl 1
Nancy H. Williams Project Manager NHW:eam
Enclosures:
Attachment A, Partial Responses to CASE Questions cc:
See ottochment 8404160053 04030V PDR ADOCK 05000445
.A.
.PDR t{
San Francisco Boston Chicago Richland
": n' :,,
March 9, I 984 Mrs. J. Ellis Attachment Response to CASE Ovestions Nicholas S. Reynolds, Esq.
Mr. John T. Collins Bishop, Liberman, Cook, Purcell & Reynolds U.S. NRC, Region IV 1200 Seventeenth Street, N.W.
61i Ryan Plaza Drive Washington, D.C.
20036 Suite 1000 Arlington, Texas 76011 Robert Wooldridge, Esq.
Worsham, Forsythe & Sampels Mr. Lo iny Alan Sinkin 2001 Bryan Tower 114 W. 7th, Suite 220 Dollas, Texas 75201 Austin, Texas 78701 Mr. Homer C. Schmidt B. R. Clements Vice President Nuclear Manager - Nuclear Services Texas Utilities Generating Company Texas Utilities Generating Company 2001 Bryan Tcwer Skyway Tower Dollos, Texas 75201 400 North Olive Street L.B. 81 Mr. H. R. Rock Dollos, Texas 75201 Gibbs & Hill, Inc.
393 Seventh Avenue Peter B. Bloch, Esq.
New York, New York 10001 Chairman, Atomic Safety and Licensing Board U.S. Nuclear Regulatory Commission 4350 East / West Highway,4th Floor Mr. A. T. Parker Westinghouse Electric Corporation Washington, D.C.
20814 P.O. Box 355 Pittsburgh, Pennsylvania 15230 Dr. Welter H. Jordon b
881 W. Outer Drive f
Reneo Hicks Ook Ridge, Tennessee 37830 Assistant Attorney Genarci Environmental Protection Division Dr. Kenneth A. McCollom P.O. Box 12548, Capitol Station Dean, Division of Engiaeering Architecture and Austin, Texas 78711 Technology Okiohoma State University Mr. James E. Cummins Stillwater, Oklahomo 74074 Resident inspector / Comanche Peak Nuclear Power Station Stuart A. Treby, Esq.
c/o U.S. Nuclear Regulatory Commission Office of the Executive Legal Director U.S. Nucit.or Regulatory Commission l
P.O. Box 38 Glen Rose, Texas 76043 Washington, D.C.
20535 Mr. J. B. George Mrs.'.S. Burwell s
Licensing Project Monoger Texas Utilities Generating Company U.S. Nuclear Regulatory Commission Comonshe Peak Steom Electric Station 7920 Norfolk Avenue Highway FM 201 Bethesda, Maryland 20014 Glen Rose, Texas 76043 Mr. H. Schmidt Mr. David H. Wode Texas Utilities Generating Company c/o Westinghouse 4901 Fairmont Avenue 2001 Bryan Tower Bethesda, Maryland 20814 Dallos, Texas 75201 Mr. David R. Pigott Orrick, Herrington, & Sutcliffe 600 Montgomery Street Son Francisco, California 94III
Comanche Peak ASLB Hearings Response to CASE Questions Question No.: Walsh //I Exhibit No.: None 1.0 CASE Question Appendix E of Cyana Report. Section DC-2.2.4. What was the yield point used for A500, Grade B tube steel?
2.0 Cygno interpretofion N/A.
3.0 Response Comanche Peak typically used a yield strength equal to 42 ksi. This may have been taken
= 42 ksi for round tubes from p. 5-214 of the AISC Code,7th Edition, which specifies Fy made of A500, Grade B, material.
The correct value for yield strength is 36 ksi, per ASME C"- c~~ vi-71-10. Cygna checked the opplicable calculations to verify that the tube steel satisfied the lower allowable. In each case, the existing design was satisfactory.
tj r
c c
I!llilllilllllitillliilitill!!.
W
Comanche Peak ASLC Hearings Response to CASE Ouestions Question b'o.: Doyle /!5 Exhibit No.: 891,894,897 1.0 CASE Question Inoccurate conclusions as related to KL/R for pinned columns:
if a column fixed at its base and free at the top has an effective K of 2.0, e
cutting of some point up from the base and adding a pin does not address the problem.
2.0 Cygno Interpretoiion For CASE Exhibits 891,894 and 897, cre the proper slenderness ratios used in calculating the column capacities?
3.0 Response The correct vobe for each component of the slenderness formula was used in ti,e evaluation of the columns shown in CASE Exhibits 891,894 and 897. This is shown in the following tables:
Exhibit 891 Used Actuoi K
2.I 2.1 L. in.
21.75 19.75 R. in.
2.99 2.99 KL/R 16 14 f /F 5%
5%
a o where K = effective length factor L = length R = radius of gyration fa = allowable oxial stress Fa = octual oxial stress l1M111111lll11111111111111111
Comanche Peak ASLB Hearings Response to CASE Questions Question No.: Doyle //5 Page 2 Exhibit 894 Used Actual K
2.00 2.10 L, in.
71.75 71.75 R, in.
!.44 1.44 KL/R 100.00 105.00 f /F 5%
6%
o o The difference in KL/R correspor:ds to a slight change in the allowable axial stress, but the actual stress remains well below the alloweble.
Exhibit 897 This support configuration transfer foods to the baseplate primarily through moments.
Therefcre, the effect of oxial column loads is insignificant.
l lillllh.illlillfillilililllll
Comanche Peak ASLB Hearings Response to CASE Questions Question No.: Doyle //6 Exhibit No.: 891 1.0 CASE Ouestion 16-inch pipe with about 20 kips load along 3-1/2 inches of length induces high bearing stresses which require pads. This is not addressed.
e ASME Code against flattening.
2.0 Cygno interpretation How did Cygno evoluote the stresses induced into the piping by the following, as related to the ASME Code caution against inducing excessive flottening into the pipe wall:
a.
U-bolt?
b.
5" x 5" x l/2" tube steel frame?
3.0 Ro w in Section I!!, the ASME B&PV Code provides the following caution:
Subsection NB-3645 (Class I Components)
" Lugs, brackets, stiffeners, and other attachments may be welded, bolted, or studded to the outside or inside of piping. The effects of ottochments in producing thermal stresses, stress concentrations, and restraints on pressure retaining members shall be taken into account in checking for compliance with stress criterio."
Subsections NC-3645 (Class 2) and ND-3645 (Class 3)
" External end internal attachments to piping shall be designed so os not to cause flottening of the pipe, excessive localized bending stresses, or harmful thermal gradients in the pipe wall. It is important that such attachments be designed to minimize stress concentrations in applications where the number of stress cycles, due either to pressure or thermal effect, is relatively large for the expected life of the equipment."
111111111111111111111111111161
Comanche Peak ASLB Hearings Response to CASE Ouestions Question No.: Doyle /,6 Page 2 The Code stoiement for Class I components is key, wherein it is specified that local effects due to flattening shall be taken into account during the stress check. It is also importoni to note that the Code does not quantify the term " flattening" for Class 2 and 3 piping. A reasonable conclusion therefore is that the designer of Class 2 and 3 piping should consider the significance of any additional stresses induced in the pipe wall due to flattening, including fctigue effects.
Asymmetric concentrated loads, capable of introducing varying degrees of ficttening, are l
common in piping systems. These loads con be produced by standard items such as clamps, frames, U-bolts and penetration sleeves.
Therefore, in reviewing the adequacy of loads introduced into the pipe wall by support l
Sl-1-002-S32R (CASE Exhibit 891), Cygna considered the following:
l a.
U-bolt. Cygna judged that the loads introduced into the piping due to design loads would not prevent the piping from performing its intended function.
U-bolts are frequently used in the industry for similor applications. To verify the correctness of this engineering judgment, Cygno subsequently performed l
on analysis of the pipe wall stresses. The results are provided in our response to Doyle Question /.'l, part d. Basically, these results show that the induced stresses are Icw and localized.
b.
5" x 5" x l/2" Tube.
The drawing details specify o O' gap of the four attachment points. Only on extraordinary construction effort could hope to occomplish this task. For example, the tube frame would have to be heated in order to shrink-fit it onto the pipe. Cygno reviewers concluded that such an effort would not occur, and therefore significant stresses _would not develop in the pipe. It should be noted that radial thermal growth would be 1/50", about the width of a mechanical pencil lecd.
- 1111111111ll111ll11111111111
Comanche Peak ASLB Hearings Response to CASE Questions Question No.: Doyle #9 Exhibit No.: 892 1.0 CASE Question The reduction of weld capacity in the calculation is based on 135 degrees. Actual tangential angle is 150.3 degrees. Therefore, an error exists. Did Cygna take note of this?
e More stress in weld than stated.
i Wide / thin ratio induces cracking as well as the 1.4:1 ratio of width to depth.
e I
2.0 Cygna Interpretation What was the basis for concluding that the stanchion-to-pipe weld shown in CASE Exhibit 892 is adequate?
3.0 Response ITT Grinnell design procedure, SA 3912,(attached) states that credit shall only be taken for the portion of the weld up to 135 degrees. Cygna concurred with this procedure and confirmed that it was properly employed on the subject support.
I l
I 111111111111111111111111111111
SA 3912 Rev. A Page 1 of 33 WELD PROPERTIES FOR PIPE /STANCHIONANDELh0V/STANCHIONCONNECTIONS FGR
.IOMANCHE PEAX PROJECT PROCEDURE SA 3912 s.
FOR INFORMATIC>N ONLY
'I Prepared By V
Checked By
[3 e,,J4 3 01,m,/,
,, f g Approved By/?/,c g,jg 4 q3jg l
Revisien A 02/08/E3 c mP-G Rc v'. /
/
- l. T. T.
GRINNELL PIPE HANGER DIVISION S A-3 912 REV. A PAGE 11CF 33 WELD ANGLES FOR STRAIGHT PIPES WITH STANCH 10N ATTACHMENTS
.Y X
A 0
@i
)
we O
A O
hi
-~
7
===.
7 b.,
s T
DETAtt. A a:
O k
y SEE DETAIL A k
-[
A
' N X
Q V
- .o 0
)
k l
THE 8 -VALUES OSTAINED FOR O VARYING FRCM O' TO 9C* ARE REPEATED EVERY 9C* FCR STR AIGHT Ph)E ATTACH M ENTS FIG. 1 g
1.T. T.
GRINNELL PIPE MANGER OlVISION s A_3912 R EY. A PAGE 20 CP 33 4.1 TABLE 1 WELD PROPERTIES OF STRAIGHT PIPES WITH STANCHION ATTACHMENTS _ (Rer. Fig. 1&4)
LIMITING WELD ANGLE = 135 NOM.
NOM OVERALL WELD PP.OPERTIES PIPE STANCH.
WELDED SIZE SIZE LENGTH Lw Sy Sx Jw Ls 1
4.24 4.24 1.36 1.36 1.79 4.24 2 1/2 1 1/2 6.24 6.24 2.84 2.84 5.39 6.24 2
8.04 5.36 4.17 1.74 7.01 5.36
.1 2 1/2 11.08 6.16 5.64 1.57 10.37 6.16 1 1/2 6.16 6.16 2.84 2.84 5.39 6.16 3
2 7.82 7.82 4.43 4.43 10.52 7.82 2 1/2 9.72 6.48 6.12 2.54 12.44 6.48 3
13.49 7.50 8.3G 2.32 18.71 7.50 2
7.69 7.69 4.43 4.43 10.52 7.69 l
4 2 1/2 9.42 9.42 6.49 6.49 18.66 9.42 3
11.72 8.46 9.29 4.60 24.32 8.46 4
17.35 9.64 13.82 3.85 39.76 9.64 3
11.34 11.34 9.62 9.62 33 67 11.34
~
6
_4 14.82 14.82 15.90 15.90 71.57 14.82 6
25.54 14.19 29.96 8.34 126.87 14.19 q
l JOR INFORMATIC N ONLY ents.
I. T. T.
GRINNELL PIPE HANGER DIVISION S A-3 912 REV. A PAGE 210F 33 4.1 TABLE 1 WELD PROPERTIES OF STRAIGHT PIPES WITH STANCH 10N ATTACHMENTS (Ref. Fig. 1 & 4)
LIMITING WELD A!!GLE = 135 NOM.
NOM.
OVERALL WELD PROPERTIES PIPE STANCH.
WELDED SIZE SIZE LENGTH Lw Sy Sx Jw Ls 4
14.57 14.57 15.90 15.90 71.57 14.57 8
6 22.14 17.22 33.86 19.76 177.62 17.22 8
33.25 18.47 50.77 14.14 279.96 18.47 4-14.46 14.46 15.90 15.90 71.57 14.46 10 6
21.65 21.65 34.47 34.47 228.37 21.65 l
8 29.05 20.97 56.44 27.95 363.95 20.97 l 10 41.44 23.02 78.88 21.97 542.05 23.02 4
14.41 14.41 15.90 15.90 71.57 14.41 6
21.45 21.45 34.47 34.47 228.37 21.45 12 8
28.40 28.40 58.43 58.43 503.93 28.40 l
10 36.56 24.37 85.53 35.49 6 50.4 6 l 24,37 12 49.15 27.31 110.95 30.91 904.37 27.31 6
21.37 21.37 34.47 34.47 228.37 21.37 8
28.18 28.18 58.43 58.43 503.93 25.18 l
14 10 35.93 27 84 89.16 52.02 7 53,37 l 27.84 F2R MFOR Y!ADCh' DNLy T
Comanche Peak ASLB Hearings Response to CASE Questions Question No.: Doyle #14 Exhibit No.: 891, 893, 894, 895, 896, 897, 898, 899, 900 1.0 CASE Question in Note 1, the same source, did Cygna consider the additive effects of self-weight excitation if the stiffness is considered from node point to hard point as opposed to the stiffness of the frame independent of hardware, local effects, base plate and anchor bolts?
Spring rate of base plate / anchor bolts (particularly bearing-type joints) con be e
considerable (observation of base plate 11 finite analysis).
2.0 Cygno Interpretation Did Cygna consider the following:
~
The effect of support stiffness on the evoluation of self-weight excitation?
a.
b.
The flexibility of each element in the support load path?
3.0 Response c.
In order to evolute the influence of self-weight excitation on support design, one must apply the appropriate dynamic loads and then calculate the induced stresses and deformations.
The applied lood, in this cose, is the support self-weight.
Support stiffness is effectively considered twice in this process.
First, it is included in calculating the opplied dynamic load. This con be illustrated by the following elementory formu!as:
1.
Load = function (freq) 2.
freq = (1/6.28) SORT (Kg/F) where freq = support fundamental frequency u
K
= support stiffness F
= self-weight g
= gravity 1N111111111111111111111111111
Comanche Peak ASLB Hearings Response to CASE Questions Question No.: Doyle #14 Page 2 Secondly, the determination of support stresses and deflections involves a structuro! evoluotion which considers the support stiffness.
For a further description of Cygno's review process relative to support self-weight excitation, see the Cygna response to Doyle Question #11.
b.
As stated in the response to Doyle question #13, Cygno recorded that support stiffness calculations on Comanche Peak were potentially deficient. When it was learned that the NRC Staff had evaluated this issue, Cygno deferred to the Staff evaluation rather than performing a redundant review.
Regarding the effects of component flexibilities on the overall support stiffness, standard practice is to consider components like anchor bolts and baseplates to be rigid, unless particularly flexible design details are used. This standard practice is both odequate and oppropriate for the following reasons:
Anchor bolts and baseplates are normally stiff relative to the overall e
- support, it is standard practice to assume anchor boltiboseplate connections as e
" fixed" in both the analysis and design. This results in a connection that is designed to occommodate the higher forces attracted by such a fixed condition.
in a standard baseplate connection, the axial stiffness for a compressive e
food would be greater than for o tensile load. This difference normally will have no significant effect on the overall piping analysis. Standard industry piping programs recognize this and have no provisions for including such refinements.
Baseplate connections are used throughout the piping system to provide e
anchorage.
Therefore, any refinements to the standard stiffness calculations would have o uniform effect on the system response. CASE Exhibit 884 shows that a large, uniform change in support stiffness has negligible ef feet on the fundamental system response.
11111111lll1118tll111111111111
P 9
Comanche Peak ASLB Hearings Response to CASE Questions Question No.: Doyle #I4 Pcge 3 The significance of individual component flexibilities located between the iood point and the baseplate connection need to be evoluoted on on individual basis.
l l
l i.
I l
Coraonche Peak ASLB Hearings Response to CASE Questions Question No.: Doyle 418 Exhibit No.: 898,899 1.0 CASE Question The base plate analysis was performed without including stiffeners alters the stiffness matrix of the base plate and consequently the distribution of moments and tension to the bolts. Beyond this point, stiffeners remain unqualified. Has Cygna addressed this?
2.0 Cygna Interpretation Did Cygno consider the bolt loading for the baseplate in the stiffened condition? Also, did Cygno qualify the stiffeners?
3.0 Response It is a conservative opprooch to ignore the effects of stiffeners on plate or bolt design, Stiffeners make a flexible baseplate behave more like o rigid piote. By making the plate more rigid, the internal moment arm created in the plate by the compressive force in the concrete and the tensile force in the bolts becomes maximum. Therefore, to resist a given opplied external moment, the mcximum bolt tension will be smaller in a rigid (stiffened) plate than in a flexible (unstiffened) plate.
Consequently, neglecting stiffeners maximizes bolt loods.
l Well proportioned st!ffeners (relatively thiuk and deep with length to depth ratio (.3) are l
generally not a problem in baseplate design. Simple and conservative stiffener analysis i
shows stresses well helow allowobles.
Detailed baseplate calculations for Sl-1-037-005-S32A and RH-1-024-01l-S22A for the stiffened and unstiffened ecses support the above observations in a general way. From the attached table it con be observed that for bolts with a larger provision ratio, the bolt loading for the unstiffened condition is greater. Bolt provision ratio is defined as follows:
1114111111111111lll11111111111
Comanche Peak ASLB Hearings Response to CASE Questions Question No.: Doyle //I8 Page 2 BP ratio =
T+
V where T = cctual tension TA VA TA= li w ble tension V = actual shear VA= Ilow ble shear it is expected that the correlation will improve with increasing bolt provision ratios.
l
[
.0 Enclosure Dl8-1 Table 1 - Support Sl-1-037-005-S32A Provision Bolt W/O W
Ratio
//
Stiff Stiff W/O W
I I,900 I,700 0.27 0.25 2
0 240 0.13 0.15 3
60 460 0.11 0.14 4
2,260 2,000 0.27 0.25 Toble 2 - Support RH-t-024-011-S22A Bolt Case I Case 2 Case 3
(/
W/O Stiff W/ Stiff W/O Stiff W/ Stiff W/O Stif f W/ Stif f I
I,170 1,580 0
0 0.40 0.43 2
I,260 800 560 610 0.35 0.31 3
0 0
1,610 1,670 0.45 0.46 4
240 770 3,140 2,930 0.41 0.45 5
3,660 2,i00 3,050 2,070 0.35 0.23 6
l 2,510 2,710 250 870 0.40 0.42
___