ML20113B623
| ML20113B623 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 01/18/1985 |
| From: | Citizens Association for Sound Energy |
| To: | Atomic Safety and Licensing Board Panel |
| Shared Package | |
| ML20113B579 | List: |
| References | |
| OL, NUDOCS 8501220048 | |
| Download: ML20113B623 (76) | |
Text
A UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION
- y; ec, BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of I
Docket Ms.Ed 1 ^45 NO :32 4
i and 50-446 n j TEXAS UTILITIES ELECTRIC 1
(Fricr
. U L__
COMPANY, et al.
I W XE I
(Application for'an' i
(Comanche Peak Steam Electric I
Operating License)
Station, Units -1 and 2) l STATEMENT OF MATERIAL FACTS AS TO WHICH THERE IS N0 GENUINE ISSUE REGARDING CASE'S 4TH MOTION FOR
SUMMARY
DISPOSITION:
TO DISOUALIFY THE USE OF SA307 AND SA36 THREADED PARTS 1.
Applicants most recently have stated that they use only A36 steel (with one exception) both in U-bolts and bolts used in Richmond inserts (see page 8 of Affidavit of Robert C. Iotti and John C. Finneran, Jr.,
Regarding the Licensing Board's December 18, 1984 Memorandum, attached to Applicants' 1/7/85 Motion for Reconsideration of Board's Memorandum (Reopening Discovery; Misleading Statement), copy attached).
2.
Applicants have admitted that to the extent that previous statements made by Applicants imply that A36 and A307 steels are identical materials, they are inconsistent with Applicants' statements in their 12/5/84 Response (Affidavit of Robert C. Iotti and John C. Finneran, Jr. accompanying ' Applicants' Response to Board Memorandum (Information on Composition of A36 and A307 Steel),' December 5, 1984) (Id., at pages 4 and 5, copy attached). d/
n /
It should be noted that CASE does not agree that this is the extent of Applicants' inconsistencies, and we will be addressing these and other inconsistencies regarding these and other statements by Applicants in other pleadings.
8501220048 850114 1
PDR ADOCK 05000445 i
Q PDR, _
s.
):
l 3.
Regardless of what name Applicants place on it, the material which Applicants have used for cinched-down U-bolts and for bolts in Richmond inserts is the same in that the material referred to as SA-307 must conform to the requirements of Specification A 36 and is therefore equivalent to, or (for all intents and purposes) is made from, SA-36 material (see attached Affidavit of CASE Witness Jack Doyle, Footnote 2, pages 3 and 4; see also Attachments B and D from Applicants' 1/7/85 Motion for Reconsideration of Licensing Board's Memorandum (Reopening Discovery; Misleading Statement) /2/).
4.
(a) The ASME Code to which Applicants are committed states, at ASME Section III, Appendix XVII, Table XVII 2461.1-1, Note 1:
" Friction type connections loaded in shear are not permitted.
The amount of clamping force developed by SA-307 bolts is unpredictable and generally insufficient to prevent complete slippage."
(See CASE Exhibit 752, copy attached; see also attached Affidavit of CASE Witness Jack Doyle, at pages 2 and 6.)
(b) SA307 material and SA36 material is unpredictable for dynamic (including seismic) loads.
(See attached Doyle Affidavit, page 2.)
f2/ It should be noted that Mr. Doyle's Affidavit was prepared prior to receipt of Applicants' 1/7/85 Motion for Reconsideration. He has not yet had time to thoroughly review it (although he did notice one particular portion which he addressed), and it may be necessary for him to supplement his Affidavit. CASE prepared this Statement of Material Facts after receipt of Applicants' 1/7/85 Motion, and we have incorporated some of the information contained in it.
From our rather brief review of it, CASE believes that the Board a ready has sufficient information contained in the codes and NRC regul:61ons upon which to make a decision regarding the issues undar discussion.
2
t 5.
The ASTM Code, to which Applicants are committed, states regarding ASTM A 307, Standard Specification for carbon Steel Externally Threaded Standard Fasteners, page 251, at section 1.3:
"Nonheaded anchor bolts, either straight or bent, to be used for structural anchorage purposes, shall conform to the requirements of Specification A 36 with tension tests to be made on the bolt body or on the bar stock used for making the anchor bolts."
(See Attachment B attached; see also attached Doyle Affidavit at pages 3 and 4, Footnote 2.)
6.
ASME specification SA-307 states at section 1.3:
"Nonheaded anchor bolts, either straight or bent, to be used for structural anchorage purposes, shall conform to the requirements of ASTM Specification A 36, for Structural Steel, with tension tests to be made on the bolt body or on the bar stock used for l
making the anchor bolts."
(See Attachment D attached; see also attached Doyle Affidavit at pages 3 and 4, Footnote 2.)
7.
Applicants' testing of the bolt material was performed to show that the bolts could take a certain static load. The test did not show whether the joints could sustain seismic loads nor what effect the non-friction joint would have on the dynamics of the system itself.
(See attached Doyle Affidavit at page 5.)
8.
(a) Applicants are using a bolting material which is only cualified for static loads.
(b) These joints present an unknown quantity as relates to the dynamics of the total system.
(See Doyle Affidavit at pages 4 and 5.)
9.
In order to cope with dynamic loads, the joints must be predictable; that is, slippage must be a controlled criteria.
(See attached Doyle Affidavit at page 5.)
3
[
I I
j 10.
The AISC Code, 8th Edition, page 5-24, Table 1.5.2.1, Threaded Partr.,
t prohibits the use of bolts and threaded materials made of SA307 and A36 l
steels subjected to other than static loads.
(See attached copy of AISC Code, 8th Edition, page 5-24; see also attached Doyle Affidavit at I
pages 5 and 6.)
11.
(a) Although in the past Applicants were only committed to the 7th Edition of the ASTM Manual of Steel Construction, the logic for the change made in the 8th Edition existed even prior to the 7th Edition if one were doing dynamic analysis.
(b) During the time prior to the 8th Edition, the AISC Code addressed l
loading in terms of static application; that is, even for structures which included earthquake considerations, the earthquake load was assumed to be an equivalent static horizontal load based on KCZW (dynamic derivation of loads was not utilized).
In the case of nuclear power plants, the earthquake loads are based on the response spectra and damping factors, in which case the predictability of the joint is required. Otherwise, both the response spectra and the damping factors are also unpredictabic.
(See attached Doyle Affidavit at page 6.)
12.
Applicants have amended their FSAR to include both the 7th and 8th Editions of the AISC Code.
(See Transcript of meeting between Cygna Energy Services and the NRC Technical Review Team, 12/20/84, page 80, lines 6 through 9, copy attached; see also attached Doyle Affidavit at j
page 6).
l l
4 I
l
.co 13.
If one built the perfect nuclear structure, perfect piping systems, and perfect pipe supports, and then connected these items with unpredictable bolting materials, one would have a total system which is no longer perfect as independent components and which is now, as a system, unpredictable.
(See attached Doyle Affidavit at page 7.)
14.
(a) The use of A307 and/or A36 threaded parts in the manner in which Applicants utilize them at Comanche Peak is a unique design feature.
(b) Applicants did not identify such unique design feature in their PSAR as required by the provisions of 10 CFR 50.34(a)(2) and (8),
which state:
"(a) Preliminary safety analysis report. Each application for a construction permit shall include a preliminary safety analysis report. The minimum information to be included shall consist of the following:
"(2) A summary description and discussion of the iacility, with special attention to design and operating characteristics, unusual or novel design features, and principal safety considerations."
"(8) An identification of those structures, systems, or components of the facility, if any, which require research and development to confirm the adequacy of their design; and identification and description of the research and development program which will be conducted to resolve any safety questions associated with such structures, systems or components; and a schedule of the research and development program showing that such safety questions will be resolved at or before the latest date stated in the application for completion of construction of the facility."
(See 10 CFR 50.34(a)(2) and (8); see also attachhd Doyle Affidavit at pages 1 and 2.)
5 L
15.
(a) Applicants recognize Messrs. Paul F. Rice and Edward S. Hoffman as authorities and attached one page (268) from a document by Messrs.
Rice and Hoffman to Applicants' 5/18/84 Motion for Summary Disposition Regarding the Effects of Gaps on Structural Behavior Under Seismic Loading Conditions.
(b) On pages 264 through 271 of the same article, Messrs. Rice and Hoffman indicate that A307 bolts are not permitted in connections subject to vibration or stress reversal.
(c) Connections at Comanche Peak are subject to vibration.
(d) Connections at Comanche Peak are subject to stress reversal.
(e) The ASME Code requires the Applicants to minimize vibration where it states:
"NF-3112.2 Design Mechanical Loads.
.. The requirements of (a), (b), and (c) below shall apply.
"... (c) Component supports shall be designed to minimize vibration."
(See attached pages 2 through 5 of Affidavit of CASE Witness Mark Walsh and Attachment A thereto, which was attached to CASE's 8/13/84 Answer to Applicants' Motion for Summary Disposition Regarding the Effects of Caps on Structural Behavior Under Seismic Loading Conditions.)
16.
ANSI N45.2.11, to which Applicants are committed, at 3. DESIGN INPUT REQUIREMENTS, 3.2 Requirements, states, in part:
"The design input requirements should include the following where applicable:
"(9) Mechanical requirements such as vibration, stress, shock and reaction forces." (Emphases added.)
17.
Supports at Comanche Peak will experience dynamic (seismic loads).
(See attached Doyle Affidavit at pages 2, and 5 through 7.)
6
- 18. Applicants cannot properly use A307 material.and/or A36 material for cinched-down U-bolts and for bolts in Richmond connections in the manner currently being utilized at Comanche Peak.
(See Material Facts 1 through 17 preceding.)
19.
For the reasons discussed herein, Applicants are in violation of:
(a) NRC regulations, including 10 CFR 50.34(a)(2) and (8); 10 CFR Part 50, Criteria I and II; (b) ASME Code,Section III, Appendix XVII, Table XVII 2461.1-1, Note 1; ASME specification SA-307, section 1.3; ASME NF-3112.2; (c) ASTM A307 specification, Note 1.3; (d) AISC 8th Edition, Table 1.5.2.1.;
(e) ANSI N45.2.11, section 3.2; and (f) standard industry practice (as indicated by Messrs. Rice and Hoffman).
9
-7
, 4..,
, ey,
o m
CASE EXHIBIT 752 4
AP=ENDIX XV*!
U*tt.2442.2-xvit.2442.5 strengtn boits m 2 hne parsfiel :o the direction of permitteo in Taote XVi!. tol.l.1 but it shall not be stress, the distance from the center of the end bolt to less than the distance spec:fied in XVil.2463.3 and that end of the con::ected part toward vh:ch :he :::e::
.eed not exceed 1% ::me: :he transserse spacing of is directed shat! be not less than A.C.'s for smc!e shear fasteners.
or '.AsC/t for double shear, where As is the nominal cross-sec:sonalarea of the bolt,eis the thickness of the XW.E We Two or Mm Fasteners la b connected east and C is the ratto of snecified of Stress Are Provided. When more than two fasteners minimum tensde strength of the bolt :o the sp' ectfied are provided m the line of stress, the provisions of X YII'2 4 2 3 Shall 0f"-
3 manamum tensile strength of the connected part.
i XVII-2M2.3 When Fastener Stress is Lower Thaa XVII.242.5 Minimum Distance to Any Edge.The j
Permitted. The end distance presenbed in XVII.
minimum distance from the center of a bolt hole to 242.1 and XVII.242.2 may be decreased in such any ed e used in desi n or in preparation of shop 5
5 propornos as the fastener stress is less than that drawin5s shall be that pven m Table XVII.2462.5 1.
TA8LE XVil 2461.11 4
ALLCWABLE SCLT TENSION AND SHEAR STRESSES Smet Sheer t F,)
Susettention Nemmes Desanonen Sett Tenseen Fe'etsen Type Benneg Tvee of True er Siae (Fel Canneetions Connesteens Feasoner Sese.
Cease eteen tn.
% of Y.5.
% ef Y.3.
% of Y.3.
Thressm9 SA.325 1
112 to I nes.
44 le t8 No -
f.I,8 to 1.t/2.ect.
50 19 19 s A.Jo 7 1 6
I I
All l
51 I
See note t t
'O Sheer SA44o 521 5
To 2 Ind.
43 16 18 pionne (Crede.VI Over 2 to 4 lacf.
42 18 17 4
To J inct.
ao 17 19 g
Over 3 to 5 inct.
l 44 17 19 3
e To J ro.
42 is la
(
Over 3 i
l-o 6 8act.
42 15 13 2
i To e snet.
I 41 1
15 17 1
1 To d snes.
i di l
15 17 S22 5
To 2 enes.
43 10 10 I
14142.He over 2 l
to 4 and.
42 to 17 4
l To d enes.
l 44 17 19 4
3 To 4 enes.
l 42
)
15 te 2
To 3 anes.
l 41 l
15 17 I
To 1 1/2 i
lect.
l 41 t$
17 i
o 023 5
l To e sacs.
p 43 le ll Is 4
1 To 9.II2 ines.
l 4e 17 I
19 3
i To 91/2 l
42 l
15 l
18
-2 l
Te 9112 41-15 17
. i~
L 1
l To 4 ines.
45 t5 17 524 5
To G Inst.
I 43 16 10 F
4 Te 91/2 44 17 19 3
Ta g.st2 tnes.
42 i,
et it 2
Te 9 8/2 Ines.
41 15 t7 t,
1 Te 8 &ne.
41 IS 17
- f. ' *": a .".. f..le 387 p*y
$i
,l*
r^
. r. i 4*"
e.r g
- \\.'
,,,[,,,,,
.===.a-
+
,j:i-o
~.1-
~
, r.;g
< e11g h,
XV11 No3-XVII.hoo SECTION ill. Div;5 TON
- SUBSECTION.u
^'s TABLE XVil.2461.1 1 (Cont'd)
Sett meer (74 Seessfesseeen Non aes Osameseen sost Tension Friction Tyse amering Tyse of Tves or Le IFt1 Cennestione Connestions Passener Sese.
Orese Clam i n.
4of Y1
% of YA
%of Ya Threesem 3A-325 I
't2 to t enes.
44
'6 24 Escsuose s.1/8 to 11/21nct.
50 19 27 SA.3o7 l 8
l All 61 l
See note 1 3Q Fienes SA44o 321 S
To a incs.
4.3 l
16 i
28 iCr.Mo.V1 4
1 To 6 incs.
46 l
17 l
27 3
To e incs.
{
42 l
15 l
25 2
To 4 enes.
l 33 l
15 l
25 1
To 4 Ines.
l 23 l
15 l
25 822 l 5 To 4 lacs.
l 43 l
16 1
26 p,
l
'4IEI #
l 4
i is e nace.
I 44 i
17 e
27 3
i To d sacs.
l 42 8
15 1
25 2
l To 31nes.
I 33 l
15 l
25 1
l To 31/2 I
lace.
23 15 23 g
823 5
l to S ind.
l 43 l
16 l
26 N##
4 l
- 3.ti2 Ince.
l 44 1
'7 l
27 3
re J 1,2 inct.
I 42 1
15 1
25 2
To 9112 f ees.
I 13 l
15 1
25 1
l To 81nce.
l 23 15 1
25 8.
824 1 5 i
fo 6 nes.
I 46 e
16 e
26 3.-
f 434o vocJ 4
1 To +8/2 ines.
l 46 6
17 l
27 3
I to s.112 taes.
i 42 15 1
25 I 2
.l Ts 31/2 Ines_
1
- 3 1
15 1
25 8 i 1
> a ines 23 1s 1
2s NOTE:
- 1. Frietsen tyse eennectione toness.in sneer are not moreistree. The senount of cronsig fores deseeooms t>v SA 307 Detts.e unersenessee. ene eenersey ensuev eent to orecent co,neeste si.eesse.
XY112463 Maximum Edge Distance XV112 665 Effective Beanng Area
},
, The maximu a distance from the center of any bolt The effecuve beanng area of bolts shall be the to the nearest edge of parts in contact with one diameter multiplied by the length in beann5, except another shall be 12 times the thickness of the plate.
that for countersunk bolts % the depth of the but shall not exceed 6 in.
counterstr.k shall be deducted.
- XVH.24H Minimuse Pitch The mantmum distance between centers of bolt holes shall preferably be not less than 3 times the SA.307 bolts, which carry calculated stress and the nominal diameter of the bolt. I.ess distance may be gnp of which exceeds 5 diameters. shall have their used only if adequate installation wrenching clear.
number increased 1% for each additional N in.in the
.7
/C) anceis available.
gnp.
= y...
sea C
S ' '.',$'
,e
? ;
d,'.
.k s
e------
c P.;. [
'65 sgj ;g NO *33 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSINdkUOARD -
u;n,fi In-the Matter of
)
)
Docket Nos. 50-445 and TEXAS UTILITIES ELECTRIC
)
)
)
(Application for (Comanche Peak Steam Electric
)
Operating Licenses)
Station, Units 1 and 2)
)
AFFIDAVIT OF ROBERT C.
IOTTI AND JOHN C.
FINNERAN, JR. REGARDING BOARD REQUEST FOR INFORMATION CONCERNING A36 AND A307 STEEL We, Robert C.
Iotti and John C.
- Finneran, Jr.,
having first been duly sworn hereby depose and state, as follows:
(Iotti) I am Vice President of Advanced Technology for Ebasco Services, Inc.
A statement of my educational and pro-fessional qualifications was transmitted with Applicants' letter of May 16, 1984, to the Licensing Board in this proceeding.
(Finneran) I am employed by Texas Utilities Generating Company as Project Pipe Support Engineer for Comanche Peak Steam Electric-Station.
A statement of my educational and professional qualifications is in evidence as Applicants' Exhibit 142B.
L A
r-
, We previously submitted affidavits regarding cinching-down of U-bolts, U-bolts acting as two-way restraints, and Richmond Inserts, which were filed in support of Applicants' motions for summary disposition of these issues on June 29, May 23, and June 2,
1984, respectively.
Q.
What is the purpose of your affidavit?
t' A.
The purpose of this affidavit is to provide information in response to the Board's Memorandum (Information on Com-position of A36 and A307 Steel), dated October 25, 1984.
Interchanaeability of SA36 and SA307 Materials O.
Are SA36 and SA307 steels the same material?
A.
No.
CASE incorrectly asserts that these materials are the same.1 Although it is true that SA36 and SA307 materials are similar, there is a major difference in the specified mechanical requirements for SA36 and SA307 steels.
As Applicants explained in our affidavit in support of Applicants' reply to CASE's answer to Applicants' motion for summary disposition regarding the effects of gaps (October 26, 1984) at 8-9, the material specification for SA36 requires both a tost for ultimate tensile strength and a 1
CASE's Answer to Applicants' Statement of Material Facts Relating to Richmond Inserts, Walsh Affidavit (September 10, 1984) at 10.
The memorandum cited by CASE (CASE Exhibit 834) is not correct in referring to the tested materials as SA307.
SA36 and SA307 rods are both used in some structural applications not involving pipe supports.
The individual who prepared that memorandum (who commonly works with those other applications) apparently did not focus on the distinction when he prepared the memorandum.
- - )
test for minimum yield point, whereas the SA307 specification requires only a tensile test (ultimate tensile strength).
Thus, unlike SA36, there is no established basis for determining certain characteristics, including relaxation, of components using SA307 material.
In summary, I
it is not appropriate, therefore, to interchange the two steels as CASE has done.
(See Memorandum at 1 ".
variability in A3G (A307) steel
.)
i Materials Employed in Applicants' -Tests in Support of Motions for Summary Disposition-
. hich tests performed in support of Applicants' motions for W
. Q.~
' summary disposition employed SA36 or SA307 steels?
.A.
None of the tests employed SA307 material in the test specimens.
Three tests utilized in Applicants' motions
~
employed SA36 steel specimens, as follows:
1)~
The tests performed by Westinghouse on cinched-down U-bolt assemblies were undertaken to1 assess the adequacy of cinched-down U-bolts to function as clamps.
In-these' tests the U-bolts.were SA36 material (see to Applicants' Affidavit on: Cinched-Down U-Bolts).
~ 2)
The tests conducted by ITT-Grinnell to determine.the ultimate capacity of U-bolts under different loading
'1 conditions, in support'of Applicants' May.23, 1984, 1 motion regarding U-bolts acting as two,way restraints, utilized SA36 test specimens.
-(see Attachment 1 tas Applicants'. Affidavit on U-bolts acting as two-way restraints.)
3)
Tests conducted by TUGCO on' Richmond: inserts in March 1983 and May 1984 were used by Applicants in support of
-their motion for summary disposition lon Richmond.In-
'serts.
(See Attachments A and F to the Affidavit supporting Applicants motion on Richmond Inserts.)
4
,2 s
RESPONSE TO NRC QUESTIONS OF MEETING OF AUGUST 8-9 and August 23, 1984 A.
U-bolt Cinching a)
Provide additional justification for the assessment that strain relaxation of the U-bolt ceases as the U-bolt stress reduces to approximately 1/2 of the yield strength.
Justification should be provided as additional data and also provide actual properties of the U-bolt material employed.
There is-scant, if any, data available on strain relaxation properties of SA-36 material.
Some relevant data is reported in ASTM DS60 " Compilation of Stress-Relaxation Data for Engineering Alloys," for material having the same composition as SA-36 steel (note that this reference does not mention the material designation).
The ASTM material specification for A-36 is presented as Exhibit A1.
Also included as Exhibit A2 are the pertinent portions of ASTM DS60 which provide data for ferritic steels having chemical composition and physical properties similar to but varying to different degrees from those of A-36.
Also provided are the definitions given in DS60, which are relevant to the question-of what'causes relaxation and whether creep is important.
Unfortunately not much data is available directly at the temperatures of interest, i.e., less than'500 F although considerable information may be inferred from the data at the higher temperatures as will be discussed later.
In fact, only materials 2 and 25 have data at room temperature.
Material 2 has the proper chemical. composition but its physical properties are significantly different from those of A-36.
Material 25 has physical properties similar to A-36 but does not quite meet al1 4
s of the chemical specifications.
Figure Al shows the stress strain _ curve of material 25 at various temperatures within our range of interest, i.e.
less than 500 F.
This curve is used to illustrate the meaning of material relaxation (as opposed to overall mechanical relaxation which will be discussed later) for mbnotonic loading, i.e. noncyclic.
For the material to relax, plastichtrainisrequired.
Ferritic steels like A-36 exhibit a well defined proportional limit at which plastic strain begins.
The yield strengths of these materials are given at the 0.1% or 0.2% elastic strain offset (in general it is the latter, although for material 25'the former is used).
In figure Al the details of the stress strain curve between the proportional limit and the yield point are not shown.
From that figure, if the material is strained below the proportional limit no material relaxation will occur.
Strains in excess of the proportional limit will ' result in relaxation, the amount of relaxation being proportional to the, amount of. plastic strain (or volume of material that has yielded).
At room temperature the strain corresponding to the proportional limit is about 0.07 5 percent.
At that level of l
initial strain, therefore, little or no_ relaxation should be expected.
Figure A4, developed using the information on Material 25isf ASTM DS60, shows that the relaxation is negligible.
At 532 F, the strain corresponding to the proportional limit point g
is 0.065 percent.
Since ttue material-25 has been strained to
(
'.075%, relaxation should be expected.
Moreover, the heating of.
'y.
the material-from room temperature-to 332 F and the return to l
l L
i 8
rocm temperature contributes to relaxation.
How this happens is explained by Figure A2, obtained via private communication with M.J.
Manjoine, one of the authors of ASTM DS60 and a recognized authority in materials behavior.
This figure is an expanded view of a portion of Figure A3, also provided by M.J. Manjoine.
Figure A3 deduces the behavior of ferritic steels like A-36 at the lower temperatures from the fact that the behavior exhibited at the higher temperatures (above 700 F) for which the data is available is the same as that exhibited for mild austenitic steels which have data available at all temperatures.
The behavior of austenitic steels is shown in figure A7 which is taken directly from reference 4 (see p. 27).
As figure A2 shows a material which is strained to or above the proportional limit will lose load at constant strain simply as a result of the lower yield strength at temperature and the higher modulus of elasticity at room temperature than at temperature.
Thus, if
' material 25 had been strained to yield at 532 F, upon its return
.O to room temperature it could exhibit 35 percent of its initial stress.
This would occur.upon return to room temperature regardless of whether " material" relaxation occurs.
If the material is maintained at temperature, loaded for sufficient time, material relaxation would also occur.
This can lead to an additional 15-20 percent loss of load.
However, for the latter time is needed to redistribute the load.
Although we do not know for a fact, it is fairly obvious that the material relaxation characteristics of material 25 at 532 F must have been determined
. at temperature, since as figure A4 indicates, there is some twenty percent relaxation.
Similar significant strain relaxation should be expected at all temperatures for initial strains of 0.225 percent, and this is indeed the case.
If the applied load results in a stress below 1/2 of the
. yield strength at temperature, the corresponding strains would be well below those corresponding to the proportional limits, and thus no relaxation should be expected.
So far only monotonic loads have been discussed.
To complete the discussion of material relaxation,.it must he pointed out that the stress strain curve for steels are different I
between the cases of monotonic and cyclic loads.
For the monotonic loads discussed so far, the point at which mild t
ferritic steel mat'erials begins to yield is higher (by approximately 15 percent - private-communication with M.J.
Manjoine) than the point at which yielding will occur under cyclic. loads.
TNe difference is shown in Figure A5.
It is important that a distinction be made between " cyclic" l
loads such as are experienced by the U-bolts, whereas the load can be cycled from a low to a.high level without stress reversal, and " stress reversal" loads which are cyclic but for which the
~
load causes the stresses to be alternatively tensile and V
- compressive.
The. relaxation behavior for the two cases can-be vastly ~different.
Figure A8 (reference 5) shows that stress l
strain curve for ferritic steel under reversing constant
amplitude loads (reversing strain).
Figure A9 (reference 6) shows an idealized curve for the kind of mild steel which is characteristic of both ferritic steels like A-36 and austenitic steels like A-304.
Figure A10 (reference 6) shows the static (monotonic) stress strain curve and the cyclic (strain reversal) curve for a material like A-36.
The cyclic curve is the envelope of the stress-strain curves exhibited during the cycling as shown by the dashed.line of figure A9.
It is important to compare the type of relaxation which one can experience under cyclic loadings with no strain reversal to those which can be experienced for the latter.
To do so we will utill:e Figure All, (provided by M.J.
Manjoine), which combines both types of loadings.
In the case of
-cyclic loading with no strain reversal, the second cycle will have a proportional limit PL1 which is about 15 percent lower than the monotonic proportional limit.
However, if the cyclic is one of relatively large strain reversal (i.e.,
strains near yield here defined as.2% offset), then the proportional limit will be much lower as indicated by point P,L2 in the figure.
?
For strain reversal conditions, according to Mr. Manjoine there is little difference between the stress strain curve of ferritic steels like SA-36 and austenitic steels like SA-304.-
Thus, the material relaxation properties of SA-36 can be inferred for-cyclic loads from those of SA-304 for which considerably more data is available.
O
Figure A6, reproduced from ASTM-DS60 (reference 4) shows the relaxation behavior of SA-304.
It can be seen that for cyclic loading with' strain reversal there can be always some material relaxation, but that for stresses below 1/2d'y, the amount of relaxation is minor.
Material relaxation, however, is only one of the parameters of interest in the overall relaration of the U-bolt assembly.
Relaxation of the assembly preload can be due to a combination of material relaxation and other mechanical relaxation phenomena that may manifest themselves during the various loading cycles, such as wear, local yielding with load redistribution, etc.
It is difficult to predict the amount of relaxation that might occur as a result of wear or yielding of surface irregularities.
It is for that reason that the long term, accelerated vibration test was conducted, i.e.,
to simulate the number of cyc,les that the assembly would see during its entire lifetime of operation.
It is possible, however, to estimate the amount of mechanical ' relaxation that takes place due to local yielding, although it is impossible to tell how quickly it wi'll' occur since the time required for load redistribution depends on "too many factors.
Such overall estimates can proceed from a knowledge of the stress state at each location of the assembly, which permits an estimate of the volume of material that might be at yield.
This volume of material will relax over time, redistributing load, and giving the appearance that the overall assembly relaxes.
It is germane to estimate what amount of m
~
relaxation could occur when the shank o3 the U-bolt is, stressed to a maximum ptress of 1/2 yield strength.
At such loads th'ere are portions, however small, of t,he assembly which experience higher stresses and can in fact be at yield.
These regions are shown in Figure A12 as points A, B,
C, D and E.-
Points A, B and C yield at the outer fibers when the U-bolt is cinched up and preloaded to belatively low value of loads as a result of straightening the U-bolt legs.
Yielding is, however, limited to the outer fibers near and opposite the pipe, and the material which yields occupies negligible volume.
For consistency with future discussion of Westinghouse test d ata, we will,use a yield strength of the material of the U-bolt equal to 36,000 psi, even though actual material yield is about 45,000 psi.
Test results obtained by strain gauges have all been referred to the 36,000 nominal yield strength.
When the stress in the shank is equal to 1/2 the yield strength in the U-bolt shank area, for instance for the 10-inch assembly (refer to l to the Affidavit) with the 3/4 inch U-bolt, the I
corresponding load is 7,956 lbs., which gives a threaded area stress in excess of 1/2 of yield, i.e.,
23,820 psi.
However, as
~ figure A13 indicates, the nut engagement results in stress
- concentration within the threaded area.
Stress concentration can I
raise the average stress above yield.
Since we have two nuts,.a similar stress concentration profile will exist in the bolt within the other nut because of the nut engagement to the first one.
For_the 3/4-inch bolt, the nuts are 5/8 inch thick with six
. threads.
Approximately half of the bolt volume within both nuts will have stress concentration in excess of 1.5.
Thus, a total length of 5/8 inches will have stresses at or close to yield.
The same is true in the other leg of the U-bolt.
- Thus, about 1.25 inches of material out of a total of 31 inches will experience relaxation of the order 15 percent (relaxation from yield stress - see figure A2) if at room temperature.
The remaining threaded area (approximately 5 inches) will experience less relaxation since it is more lightly stressed.
The amount of relaxation that it can experience can be estimated using figure 2,
suggested by M.J.
Manjoine.
This additional threaded material would relax approximately 7.5 percent.
Thus, one can approximate the overall mechanical relaxation that would occur for loads resulting in stresses in the shank of one-half yield as 5 (.075) + 1.25 i.15) = 1.7%, or very low relaxation.
3.25 Perhaps more relevant than theoretical calculations to the t
(
question of when overall (material and mechanical) relaxation ceases for the U-bolts, is the actual data taken during the various tests conducted by Applicants (see reference 1).
One i
i such-test is the thermal cycling test.
L Results of the thermal cycling test on the 4" Sch 160 stainless steel specimen indicated that the stress in the U-bolt was approximately 31,100 psi (or approximately 86.4% of the assumed yield strength of 36,000 psi and essentially equal to the cyclic yield strength).
The total material'would thus relax.
~
1
_9_
After nine cycles the residual stress was measured to be approximately 19,900 psi or 55 percent of the assumed yield strength.
( Ambi.ent temperature for pipe and U-bolt was essentially the same before cycling (105 F) and just before the 10th cycle (107.5 F).
The U-bolt was heated to an average 0
temperature of about 400 F (see page 16 of Attachment 3 to the Affidavit).
From Figure A2 one can deduce that the temperature cycling would result in a relaxation of approximately 36 percent, of which the initial 25 percent would be due to the temperature cycling.alone.
The result of the thermal cyc.' 8.ng test does in fact confirm that the room temperature stress before the thermal
- cycling, i.e.,
a nominal 31,100 psi, was reduced to 19,900 or a 36 percent reduction.
Another test which provides insight on the stress relaxation is the creep test which was performed immediately after com-pletion of the thermal cycling test, without retorquing the bolts.
For the 4-inch specimen the microstrain measured in the two U-bolt legs at the ambient temperature before the creep test (77 F) were 856 and 775 microstrain for legs 1 and 2 respectively.
(These microstrains correspond to a load of 4,870 and 4,409 lbs.)
After the creep test with the ambient temperature being 91.4 F, the strains were measured to be 853 and 773 microstrain, respectively.
When one accounts for the fact that' at 91.4 there is a preload induced by the difference in thermal expansion between the stainless steel pipe and the carbon
steel U-bolt, and that had the ambient temperature returned to 77 F the preload would have been reduced by approximately 45 lbs., the final, load at the completion of the creep test would be approximately 4,580 lbs. compared to 4,639 (or 1.2 percent decrease).
Since 4,580 lbs. corresponds to a stress of 23,367 psi (shank area), which is above 1/2 of the assumed yield strength of 36,000, this decrease, if real and not due to instrument uncertainty, would be due to the strain relaxation.
The question-of whether it may be due to creep is addressed in the answer to the next question.
For the 10" Sch 40 line, where the temperature is low (pipe 250 F and_U-bolt 150 F) creep is clearly not a concern.
The strains measure prior to the creep test (after the thermal cycling test) were 283 and 280 microstrains respectively in legs
~1 and 2 of the U-bolt (at an ambient-temperature of 75.8 F).
The initial microstrains correspond to a load of 3.625 and 3,578 lbs.
respectively.
These loads ' correspond'to a stress equal to 8,200 psi in the shank or 10,800 psi in the thread area of the U-bolt.-
In either case the stresses are well below the-1/2 yield strength, with the exception of highly local area in the ' thread within the nut, and hence little,-if any, relaxation should be exhibited.
The strains after the. creep test were measured to be 281 and
.276 microstrains respectively corresponding to an average load of
~
3,567 lbs.
+
- 11 The drop in load of approximately 39 lbs. is partly due to the lower environment temperatures after the test which was 66.9 F instead of 75.8 F.
The drop in load corresponding to the 9 degrees difference is calculated to be approximately 11 lbs.
Thus, relaxation (if any) was less than 0.8 percent.
The seismic test provides further evidence of the relaxation phenomenon.
Initial information provided from the test, which is attached as Exhibit A3, indicated a reduction in load from 4,484 lbs. in both U-bolt legs to about 4,291 lbs. and 4,355 lbs. in legs 1 and 2 respectively, when the assembly was vibrated at 9 Hz with a constant amplitude of 7,000 lbs.
This relaxation of approximately 12 percent could not be justified on the basis.of-the applied load which would result, coupled with the initial preload of 4,484 lbs. (50 ft. Ib. torque) in maximum load experienced by the U-bolt of approximately 6,100 lbs., and a corresponding stress of 18,200 psi in the threaded area and 13,800 psi in the shank area.
This led to questioning the validity of the 7,000 lb. load, and to the realization that the actual applied vibratory load had been higher, and to the results
-published in the Affidavit, which are included here as Exhibit
-A4.
As-seen in the Exhibit, the actual load applied to the U-bolt was in excess of 10,000 lbs Lduring the peak portion of the cycle and initially in excess of 8,600 lbs. during the pull
- portion of the cycle.
On the average. the force seen by the U-bolt during the cycling was in excess of 6,600 :1bs. (peak load of
s more than 8,600 lbs. plus preload of 4,484 lbs.) which would have resulted in a stress in the thread area of about 19,800 lbs.
which is 11 pere.ent higher than the nominal 1/2 yield strength, hence justifying the relaxation seen.
Finally, the data obtained during the long term accelerated vibration test merits some attention.
As stated in our Affidavit, the initial preload stress was equal to about 9,020 psi.
After the initial reposition of the assembly which occurred approximately 5.15 minutes into the test (see attached raw data - Exhibit AS), and which resulted in an average loss of preload equal to 640 lbs, the preload was seen to decrease slightly, then-increase again then decrease with a final preload being about 450 less than the preload existing after the initial adjustment.
During the period of time between the 4th sweep (21 minutes) and the 36th sweep (189 minutes) there was essentially no. change in the preload.
At the latter time is when the sudden cocking mentioned in the Affidevit on p.
30 took place, which resulted in some further preload decrease.
Relaxation of the material discussed within the context of this reply does not change the total strain of the material.
(See definition in 2 of Exhibit' A2.)
The preload at the end of the test is still sufficient to prevent loss of contact between the pipe and backing plate (see figures 17 and 18 of Attachment 1 to the Affidavit with an applied load of 1,500 lbs. and a preload of approximately 3,200 lbs.), thus the motion which resulted in further relaxation is most likely due to accumulated strain over
=
t,
6 the more than 10 cycles experienced at an applied load of 1,500 lbs.
These cycles represent the number that the support may experience during its lifetime, and hence the test results confirm that in spite of some relaxation, adequate preload would
~be retained throughout life.
Cyclic plastic strain accumulation may occur at these loads, which are abnormally high for the period of time tested.
An elasto-plastic finite element analyses of a similar U-bolt, backing plate, pipe arrangement, conducted per an 8-inch pipe (same size U-bolt as'the 10" pipe, indicates that for sufficiently high proload, the U-bolt can experience come
. plasticity in the transition region between the straight. shank and~the curved portion and at the inner surface of the U-bolt apex.
This occurs from the. bending moment place on the U-bolt from the straightening action of the preload or full external L
. load.
This small amount of plasticity occurs even though the L
average stresses through the U-bolt cross-section is low, and in L
fact,Ifor the'particular case examined are only 2,000 psi.
Under I.
the large number of cycles seen by the specimen the accumulated plastic: strain can result in sufficient permanent deformation to permit. relaxation.
Also, wear and yielding of surface-l imperfections can accomplish the same thing.
f
.. b)'
Provide more-information as to why creep of the U-bolt l should not be; a consideration, considering the result for. the 4-inch pipe.- Provide material of U-bolt nut.
Include explanation on effect of different enbient temperatures on loss of preload l--
shown by_this test.-
i-i_
r
. Relative to the possibility of creep phenomena existing in the U-bolt, the maximum temperatures measured for each of the three test specimens, during the Creep Test are listed below.
It is to be noted that the temperature in the U-bolt varied along its length.
4-inch specimen Pipe temperature:
560 F U-bolt temoerature:
445 F
^
Nut temperature:
340 F 10-inch soecimen Pipe temperature:
250 F U-bolt temperature:
150 F Nut temperature:
140 F 32-inch specimen Pipe temperature:
560 F U-bolt temperature:
350 F Nut temperature:
170 F Also note that all three U-bolts are SA-36 Carbon. Steel.
Reference 2 suggests a temperature of 752 P (400 C, 673 K) as the
. minimum used for creep tests performed for carbon steels.
Finite creep is not discernable in carbon steels at temperatures lower than this.
Figure A7 (from reference 4) further confirms this.
Reference 3 defines the temperature below which self-diffusion is too slow to influence creep as approximately one-half of a metal's absolute melting temperature.
The absolute melting temperature--for SA-36 carbon steel is in excess of 1366 K (1093 C, 2000 F).
Similarly, reference 4 defines the temperature below-which creep is not discernable as 0.4 Tm (Tm metal absolute melting temperature) which would correspond to 524 F.
L.
7 Based on the fact that none of the U-bolt temperatures exceeded 500 F it can be concluded that no finite creep occurred in the U-bolts..Since the nut material is ASTM-A563GrA and none of the temperatures exceeded 340 F, no finite creep occurred in the nuts.
The curve shown in Firgure A3 for ferritic steels like SA-36, and Figure A7, taken from reference 4 for austenitic steel, confirms that relaxation is not due to creep until temperatures of approximately 800 F are available.
The su.ll decrease in U-bolt preload experienced during the test, of the 4 inch sch 160 pipe is believed to be a result of relaxation as explained in the answer above.
Based on the above, and test results obtained, it is concluded that none of_the U-bolt test specimens were subject to creep phenomena during the Creep Test.
The explanation of the effect on the loss of preload from the different ambient temperatures is given in the answer to the preceding question, namely the higher ambient temperature at the end of the test would have the effect of underestimating the loss of preload-by about 45 lbs.
c)
What is the thickness of the backing plate,for the 4"
-pipe - U-bolt configuration?
The thickness of the backing plate is 3/4 of an inch.
The drawing provided was a poor copy where the copying has resulted in a 3 looking like a 1.
Enclosed (Enclosure A1) is a better copy of the drawing reflecting the 3/4 inch thickness.
- d)
Clarify the statement made in the opening remarks regarding the 32" pipe on page 42 of Attachment 3 to the Affidavit.
The statement as written:
"The stresses measured in the test and calculated for the 32" pipe, cross piece and U-bolt are comparable." was not meant to state that the magnitude of the stresses calculated or measured were comparable numerically.
It is quite obvious from the 32 inch pipe test data that the data scatter would make such comparison questionable.
It simply meant that the very low stresses calculated by finite element analysis were confirmed to be low by test.
e)
Verify that stresses in the pipe would still be acceptable if one had used the C indices rather than the B indices of the-Code on p.
54 and following of the Affidavit.
This question refers to the effect on the pipe stress intensities that would be computed, had the piping moment stresses been computed utilizing the C indices (Class 2 and 3) rather than B indices (Class I).
The effect of ASME Class 2 and 3 rules on the piping stresses has been discussed in the affidavit on pages 63 to 66.
On page 65 of the affidavit, a comparison is made in Tables L and M of the deadweight and seismic (Equation 9 - Class I rules) and the thermal (Equation 12 - Class 1 rules) piping moment stresses developed using Class 1 and Class 2/3 stress indices.
~
. The changes in stress indicated by the results reported in 1
these tables are given below (Table A ).
Note that a positive value implies an, increase in stress, and a negative sign a decrease in stress if Class 2/3 rules are used.
l TABLE A CHANGE IN PIPE STRESS CHANGE IN STRESS (KSI)
DEADWEIGHT +
PIPE SIZE MATERIAL SEISMIC THERMAL 4" SCH 160 Stainless 2.15 4.01 10" SCH 40 Stainless 2.55 0.07 10" SCH 80 Carbon 1.97
-2.86 32" MS Carbon 2.49
-0.32 The results of this change on the stress intensities calculated 1
using Class 2/3 rules is given below (Tables B and C1).
These tables can be compared to Tables H and I given on page 60 of the Affidavit.
e:*
9 6
18 -
TABLE B TOTAL PIPE STRESS INTENSITY PRELOAD APPLIED TOTAL STRESS PIPE SIZE TOROUE STRUT LOAD INTENSITY 4" SCH 160 60 ft/lbs 2,000 lbs 70.3 ksi 10" SCH 40s 100 ft/lbs 10,000 lbs 7 6.83 ksi 10" SCH 80 100 ft/lbs 10,000 lbs 53.80 ksi 32" MS 240 ft/lbs 100,000 lbs 49.34 ksi TABLE C MAXIMUM PRIMARY AND SECONDARY STRESS INTENSITIES EO. 9 EO. 9 EO. 12 EO. 12 PRIME STRESS ALLOWABLE SECONDARY ALLOWABLE PIPE SIZE INTENSITY (KSI)
(KSI)
(KSI)
(KSI) 4" SCH 160 33.75 50.52 36.55 50.52 10" SCH 40S 63.16 60.00 13.67 60.00 10" SCH 80 40.12 60.00 13.68 60.00 3'" MS 33.06 58.26 16.28 58.26 With the exception of Equation 9 for the 10" SCH 40S pipe size, all of the pipes evaluated meet the_ Equation 9 and Equation 12 allowables.
The Equation 9 stresses reported for the 10" SCH 40S pipe are conservative since:
1.
The pipe stress includes the secondary stress due to pressure pipe growth restriction.
.. 2.
A higher stress push load is used than seen by the Comanche Peak 10" U-bolt supports.
3.
A higher mechanical primary pipe moment stress is used than seen by the 10" Comenche Peak pipes.
The significance of each of these items is given below:
1.
The total circumferential pressure stress from the computer analysis is 10.51 ksi.
The circumferential pipe stress due to pressure is 8.84 ksi.
The secondary pressure stress is'10.51 - 8.84 = 1.67 ksi, which is presently included as primary stress.
2.
The largest U-bolt strut load as determined from ITT Grinnell U-bolts loads is~8,585 pounde.
In the evaluation, a 10,000 load was used.
This is equivalent to a 2 ksi reduction in pipe stress.
(72.71) - 58.59 [1- ( 8585 ) ] = 2.0 kai
=
10,000 3.
On pages 61 and 62 of the affidavit, a comparison is made between the primary piping moment stresses used in-the U-bolt evaluation to actual randomly selected computer piping analysis stresses.
From Table J of the affidavit, it can be seen that the mechanical primary pipe moment stress used in-the U-bolt cinching evaluation is 3.3 ksi higher, (10.45 - 7.063 = 3.3 ksi).
_j
. Adjusting the 10" SCH 40S stress intensities given in Table C to remove the conservatisms discussed above results in a primary stress intensity,value of 56.19 ksi.
Note that without consideration of item 3, 3.3 ksi, the primary stress intensity value is 59.49 ksi and is still below the allowable stress.
The secondary stress intensity if 15.34 ksi.
Thus, the 10" SCH 40S pipe is within the acceptable limits of 60.0 ksi for primary and secondary stress.
f)
Provide an example of how the total value of stress intensity can be obtained from the finite element results and how the value can.be divided into equation (9) and equation (12) stress intensities.
The easiest way to show how the stress intensity is obtained is to refer to the figure VII-2 of Attachment 3 of the Affidavit which defines it as the maximum of either the absolute difference between'the major principal stress or minor principal stress and zero or the algebraic difference of the two principal stresses, and to apply'this figure to an actual example.
The example i
chosen is the 4" sch 160 pipe.
For the elements having the L
largest circumferential and longitudinal stresses, the finite I
element analyses determined 1that the principal stresses are virt.11y identical to the circumferential and longit'udinal stresses (see Attachment 3 of Affidavit at page 57).- The longitudinal, circumferential, major and minor principal stresses t
for the highest: stressed piping element of the 4" sch 160 pipe i
l
, are given for both the inside and ot tside surfaces and for the maximum load case in the table of p. 58 of Attachment 3 to the Affidavit.
These values are reproduced below:
Long.
Circum.
Princ. Stress (ksi)
Stress Stress Major Minor (ksi)
(ksi) 4" sch 160 inside 10.49 44.79 44.78 10.50 outside
-26.65
-34.07
-26.63
-34.08 where the negative sign denotes compressive stresses.
A confirmation of the max. circumferential stress can be found in the table of page 71 of Attachment 3 of the Affidavit for element 627.
Note that on that table, there is no distinction regarding the surface at which the maximum stresses occur.
For instance, the 44.79 kai tensile circumferential stress occurs on the inside surface, while the -26.65 ksi compressive longitudinal stress occurs on the outside surface of element 627.
To the local stresses computed by the finite element analysis one must add the longitudinal equation 9 pressure and pi. ping moment stresses.
These are available from the table on page 56 of attachment 3 of the Affidavit.
They are:
b
. Longitudinal Pressure Stress 4.8 ksi EQ. 9 Piping Moment.
2ss
+ 12.146 ksi EQ. 12 Piping Moment Stress
+ 22.49 ksi Adding the longitudinal pressure to the stresses previously tabulated we obtain:
Principal Stresses
!!ajor Minor
( Circum ferentia l)
(Lonaitudinal) 4" sch 160 Inside 44.79 15.29 Outside
-34.07
-21.85 To add the piping moment stresses to the longitudinal (minor principal) stresses, we choose the sign which will produce the largest stress intensity.
This-is seen in a Mohr circule depicted in Figure A14, where inside surface stresses are used.
Thus, the total stress intensity is given by 44.79- (-19.346)
=
64.136 ksi, which is the total stress intensity given on page 59 of Attachment 3 of the Affidavit or in table H of page 60 of the Affidavit.
For comparison purposes, the stress intensity derived for the outside surface is:
Maj. Princ. (Circumferential) stress = -34.07 Minor Princ. (Longitudinal) stress = -26.63 + 4.8 + 12.146
- 22.49 = -567466
1 The max. stress intensity is thus 56.47 ksi.
Using the alternative signs would have produced a stress intensity of 34.07 + 12.8 = 47.5 ksi which is lower.
As snown above, the highest stress intensity occurs on the inside surface.
To determine the primary and secondary stress intensities, several alternatives are available.
The most straightforward determines the primary stress intensity from the principal primary stresses and derives the cecondary stress intensity by subtraction of the primary from the total.
For the example chosen, we proceed as follows:
(i)
The secondary portion of the circumferential stress is obtained as the stress due to thermal expansion by subtracting the'circumferential stress due to preload
+ thermal given on page 59 of Attachment 3 of the Affidavit as -39305 psi, from the circumferential l
stress due to preload alone, which is given in the preceding page as -26c91 psi.
These occur on the I
outside surface.
The primary circumferential stress l.
becomes -34.07 +13.21 = -20.86 ksi.
[
(ii)
The primary longitudinal stress is similarly derived by considering only the equation 9 piping moment I
- stress, i.e.,
neglecting the equation 12 stress and l
subtracting the difference between the longitudinal l
t
. stress due to preload + thermal and that due to preload only, which equals 6.5 ksi.
The longitudinal stress thus becomes -21.85 -12.146 + 6.5 = -27.5 ksi.
.(iii) Thus, the primary stress intensity is -27.5 ksi and the secondary stress intensity becomes 56.47 - 27.5 =
28.97 ksi.
Similarly, we obtain the primary and secondary stress intensities for the inside surface.
(i)
' Primary circumferential 44.79 - 10.81 = 33.98 (10.81 is the difference between preload + thermal and preload_only circumferential stresses for the inside surface and these do not appear in any table, but are available from the computer output).
- ( ii)
Primary longitudinal = 15.29 + 12.146 - 4.24 = -1.1 where again 4.24 is the difference between the longitudinal stress due to preload + thermal and that-due to preload only.
Please note that the primary stress intensity is thus 35.1 ksi instead of 'the value of 31.6 reported on page
'59 of the Attachment 3 to the Affidavit.
(iii)'The secondary stress intensity then becomes 64.14 -
- 35.1 = 29.04 ksi instead of the 32.54 ksi_ reported.
The: difference!between the numbers here and in the Affidavit-occurred When inadvertently the outside secondary circumferential
-stress was subtracted from the inside total circumferential stress.
k
O.
g)
Define what is meant by partial preload in the tables to the Affidavit.
Partial preload refers to a loading condition in which the torque of the U-bolt is a fraction of the maximum torque that is assumed to be applied to the U-bolt.
For instance, for the 4" sch 160 pipe U-bolt assembly full preload corresponds to a torque.
of 60 ft.-lbs.,
and partial preload corresponds to a torque of 9 ft.-lbs.
'h)
Confirm the location of strain gauges SS and S10 in Figure 21 of the Test Report SQ&T-EOT-860 (Attachment 1 of the Affidavit).
Sketch 5 on p. 58 of Attachment 1 of the Affidavit is in error.
It inadvertently suggests that the same U-bolt strain
. gauge identification scheme used for the Torque vs. Preload, Friction and Load Distribution Tests was used for the Thermal Cycling and Creep Tests.
This was not the case.
Since high temperature strain gauges were required for the Thermal Cycling and Creep-Tests, low temperature gauges that may have been used for previous tests were removed.
The high temperature strain gauges were not instrumented to be consistent with the low temperature gauges.
Also, the low temperature gauges were identified by BLH channel number.
When test data for Thermal Cycling and Creep Tests was first received from the lab, the strain gauges were identified by serial number.
Thus, in EO&T-l EQT-860, the strain gauges used for the Thermal Cycling and Creep Tests are not identified by the sample Si through S5 sequence as t
1 o
in the other tests.
.Since channel numbers are directly traceable rb' port are to serial numbers, any results contained in the test easily traceable,to the appropriate test data.
The high temperature strain gauges as installed for the Thermal Cycling and Creep Tests are identified on the attached Figure A15 for each of the U-bolt sizes.
The strain gauge on the three-gauge U-bolt leg that is located 90 from the two other gauges is not required to monitor U-bolt preload and, therefore, is not referenced in any of the test results.
(i)
Correct typo on p.
66 of' Test Report Leg 2 (gauges S4, S11) should read 3516 instead of 5316 pounds.
(j)
Provide material properties of the U-bolts and nuts used.
The mechanical properties of the.U-bolts are as follows:
1/2"_U-bolt Sy = 45130, 45290 psi; Su = 63080, 63590 psi 3/4" U-bolt Sy = 44350; su = 65120 psi 2 3/4" U-bolt Properties not provided by Vendor Nuts ASTM - A563 GrA.
i l
l l
L
References:
1.
Report No..EQbT-EQT-860, " Comanche Peak Steam Electric Station U-Bolt Support / Pipe Test Report".
2.
I.A. Oding, " Creep and Stress Relaxation in Metals", Oliver and Boyd, 1965.
3.
F.
Garofolo, " Fundamentals of Creep and Creep-Rupture in Metals", Macmillan, 1966.
4.
Compilation of Stress Relaxation Data for Engineering Alloys, " ASTM Data Service Publication DS-60."
5.
- 4,
- p. 275, 1973.
6.
H.R.
Jhansale and T.H.
Topper " Engineering Analysis of the Inelastic Stress Response of a Structural Metal Under Variable Cyclic Strains," ASTM STP 519, 1973, pp. 246-270.
W
'e
l 3.... : -
1
~ ~,
t 6556 i
?
?f di a:
>~.
MR. VIVIRITO:
One moment, please.
Tb
.+
If you look at the test result, and you are
}l i.~
l'"
concerned with the question of shakedown and twice yield, di'
,j when.you look at the test that tests the entire assembly,
[
t the bolt, the Richmond and the concrete, there really is no pronounced yield before you go to ultimate.
'., g pf 1
l
-l It is rather gradual.
,M Mhen vou look at the typical test strain curve I
3:?
for steel, you have a very sharp yield for a lona point, 94 ff:a-13 i
and then it coes up.
The assembly of the entire bolt, the bolt, the Richmond and the concrete, there is no 5. '[,,.
real pronounced yield.
So it really is not relevant i
- j-
.l L 1 in terns of shakedown.
?,
JUDGE BLOCH:
What you are saying is that the
' A..,
yield for these bolts at the time that Stardyne was
-k:-'n to run was lower than it should have been?
i f
5_
17 MR. VIVIRITO:
No, sir.
What I'm saying is
.m I-s.a 18 that the entire assembly must work together.
You have p.:
9:-
19 the bolt and the Richmond and the concrete.
And the only
+
- (.
20 way you can evaluate the entire assembly is by a test, e
it 21 and indeed the observed phenomena in running these 2.,a-tests is that there is no real -- there is no real
- 9 22 y
- .e ec 23 pronounced yield.
7,..
7 24 JUDGE BLOCH:
There is no yield on this
,u 4
2.
25 bolt?
a E[
U s
,m h-n Xi
'RL
a.
6557 i
4' t.
1 MR. VIVIRITO:
1:o, not on the bolt.
On the 2
entire assembly.
2 F
l 3
What we are looking at is the way the entire l
/
4 assembly works.
And you can't really analyze the j
v 7#.
5 interaction between the bolt and the Richmond and the 3
concrete.
That is why a test is the only valid way 4.
~
7 to arrive at an allowabic.
7
..a i
And when you look at that entire assembly, i
l l
inceec when you look at this curve as compared to
'3 a stress strain curve for typical steel, stress strain
'l curve for typical steel has a very pronounced curve 1 12 for a long time, and if indeed you go too far you ]71' ~ 13 get a considerable elongation because you are in the it 'd fla t part of the curve. I is These curves do not have any pronounced 'o yield. And.that is the phenomena of a steel concrete 17 interaction. f 18 JUDGE BLOCll: So at the very least this 19 unit acting as a whole has a yield that must far 4 . k. O exceed normal stresses? k. 21 That is based on this test? .l 22 MR. VIVIRITO: There is no pronounced yield.
- k 23 It goes finally to an ultimate.
24 JUDGE BLOCII: No' pronounced yield also means .z 25 that the yield is far above the normal stresses. q 9 i s
ATTACHMENT D SPECIFICATION FOR CARBON STEEL EXTERNALLY AND INTERNALLY THREADED STANDARD FASTENERS SA-307 (Ideaucal with ASTM Specancaemn A Jo716e except for Pers. 2.s and 4.1) I, Seepe 2.6 Galvanized bolts, nuts and washers 1.1 This specification covers the chemical shall be hot-dip galvanized in accordance with and mechanical requirements of two grades of the requirements of ASTM Specification carbon steel caternally and internally threaded A 153 Class C forzinecoating(Hot Dip)on standard fasteners, in sizes 14 in. (6.35 mm) Iron and Steel Hardware. Nuts ahall be tapped through 4 in. (104 mm). This specification oversue, after salvanizing, by the diametral does not cover requirements for esternally amounts listed in ASTM Specification A 563 .-threaded fasteners having heads with slotted for Carbon Steel Nuts unless otherwise or recessed drives. The festeners covered by specified. When specified by the purchaser, this specification are frequently used for the fasteners may be mecharucally galvanized following applications: 1.I.I Grade A Aalss, for general applica. provided that the costing and consed product i meet the coating thickness, adherence and 1.1.2 Grade # Achs. for flanged joints in 9 Y "9"
- * " * '^
i piping systems where one or both flanges are A 153. Class C. MechanMy salysanzed nuts for assembly with mechanically galvanized 1.2 If no grade is specified in the inquiry, bolts shall be tapped oversize prior to mechan. contract, or order. Grade A bolts shall be Ically galvanizing but need not be retapped furnished. after mecharucally salvenazing. Nota shall be 1.3 Nonheaded anchor bolts, either straight provided with a suitable lubricant, or bent, to be used for structural anchorage purposes, shall conform to the requirements
- 3. Chessical Requirensees of ASTM Specification A 36 for Structural 3.1 Steel for bons and nuts shall conform Steel, with tension tests to be made on the to the following chemical requirements-bolt body or on the bar stock used for making c,,4, A o,eee a L
the anchor bolts. Non-The values stated in U.S. customary "'.0s**"."12 i units are te be regarded as the standard. Pheesterms.maa.t 0.06 0.13 0 0 seirer. mas. s 0.13 0.23 0.05 0 15
- 2. Mosenals and Manufactum 2.1 Steel for bolts shall be made by the
- II"
- ***N *' " " " ".
open. hear.h, basic.onygen, or electric. furnace Mi'Ct 8 888 08p
- M M F" M 3.3 Bolts and nuts are customarily fur.
'""*****I"* Steel for puts shall be made by the l open-hearth, basic.ozygen, electnc. furnace, or heats of steel cannot be identified. bessemer process. ~ 3.4 Application of heats of steel to which 2.3 Bolts may be produced by hot or cold bismuth, selenium, tellurium, or lead has bee n forgmg of the heads or machining from bar intentionally added shall not be permitted stock. for Grade B bolts. 2.4 Bolt threads may be rolled or cut. ' 2.5 Nuts may be produced by het pressing.
- 4. Mechanical Requirements cold penching, cold forging, or machining 4.1 Bolts shall not exceed the maximum
.from bar stock. hardness required in Table 1. Bolts less than 339 I y6-m s. .m ~ + ~ -.---- -. - - < - - -, - - ~ ~ - - - - - - * - - - - - ' - - - - ' - - * - - ~ ' ' ' ' - - " ~- ' ' ~ ~ ~ '
i I .e. ATTACH:ENT 3 t-1 4 l ^ l if l tt t ; f. ~ .j ' 4 fe.w am.cc W i w Y=.;;~n and . ASTM A 307 - 80 t j ,f! '.I i I t Standard Specification for CARBON STEEL EXTERNALLY THREADED STANDARD FASTENERS' N mera.s n, ca i.ncer,e t sea ces enu en 4 r er ee mmea.xeai c..r e ne e n,.umn,n.:aes e.e . ear -' 7: 4. a a O " '? Or l'1 I*! sake cl Te's 'Wn l'e % ear ( l.4.31 *e.1%.un A %rn-f r m Care? tr. eses artdaa;cs trc g ea r cf.au rearrrein t l r l 1..< cope 1: The unues stated m inch-round unas i Inn specua rion cas ers rne encrmc.a are a ne regarcea a3 me uancard. k j and meenanica recutremems et two crudes of
- 2. \\pplicable Docurnents can on s ecI esternaih inreaced s:and..rd f..,
\\ h T't b:..na..cas tv i tenert in s:zes a in in 4 mm i tnrouen, in 1; . I m. mmi
- ha spec tication Joes not toser
.s.a srec:: ca::an f or M ue:ur.n Yee: ejun e* J"!, for es,T h.s inreJJed l D ris n irm.. na 5:t.: I L J.s.. I i en;, na me ne as acn .ateJ or s. ecd
- ..' \\fe
- nods.nJ &7!:rntmrn ;or \\fe
. i ws or :or metnan:cai eeansion ncnars i' ,,ine !.oteners cosereJ rs tna spec:n;;:ron an nan:cm! Testme or steel Proaucs A m.. 5pec:: c.tmn f or Carron and \\l;os +. "ccaent. usca far Inc :odowine appucations ,1 I Graac.4 llom. or cencrat appaca-5:ce! Nuts F B ~.5J 5cccirica:ron far %fecnanicam De o tions. nd [, l !., Graae B llom. :.or t'anced tomts m posued L,oatines or CoJmium.;na L:nc J on i.erraus \\le:R p
- ip
- ne mtems wnere one or Ntn :.ances are at.ma 8 ta n ta,,
cast iron ... kr tran V. F i.
- f. if
,. If ro traJe n srec:: cc in the maurrs \\%f bl ' I.muca $;*ew Th -ads t it A %! Hi * ! Sauare..nd Hex tiods co n t ra c. -. i r e. Grade Am...- .h a n. r. ff o o
- urn m ec screws 1
NinneJJed ancnor roih. cither straicnt
- 3. \\IJterials and \\lanuf.nture w
or rent. :a re used far structural anchorae: 3.1 9 eel for noits shali he made hs the purposes. shad conform to the reauirements at %eaf ram \\ % witn rension tesa io re open.heartn. mic-ou ren or ciecmc-rurnace ~ process. mac on rv "oir noos or on tne rar stocs d used for maune the ancnor rolts. 14 Suitable nuts are cosered in Spectftca- - Thn snee 6 canon is un.2er the u.nsaicnon cf AsTsf i Co.,m.ee n n, n., n ue,en a.ia n u.e area resp.mu. tion A 5h3 Unless otnerwac specified. the W, of Secumnee F 16 t'2 on vees hwe craae and srsle of nut for each erade of Carfent edition approved Acr.: 25. IN Pubusnea Lne He oneina.iv ., fastener. of all surtJCe finishes. shail be as eeen a m,ruonsned aa A M - 47 T Lau previms ~ k 0* 8 For A WE hooer and P< enure Vewel Co:e arrua- 'Ai Fauener GraJe and Sac Nut Grace ana stue. Co2e tmns.ee reia:ca $peancaton 3 A 307 in sectian 11 et ma f. A. . #o l '/a an A. be s ^ 4"nw But of 4 S TV %:andarar Part e
- 4 mu B w at 4 % TW hanaaras. Part )
~ A i ver la to 4 in A. hean bes $.4 anu Bane at 4 STV %.anuaru. Parts I and 4 B. '. to 4 in A. heen het
- 4new Bw of 4 VT W 5:awarst Parts 4 a,a 9
- Nuts n. ather graaes and stvies havine steched prmf
' 4".u.as Bm at AS TM Siamaras. Parts a 2. 3. 4. 2 and 10 1.iad streurs e specticanon A 54. Tame 3) it ester than 6 a stav he netoned from Arrenan Natteral %nturm ' ; the srects ed race and uvie ut riut are a.so swtanee. Irstitute. loc.,14 3u broadeav. See i vra. N i tuols
- y..
M t- \\: 'Ls, t 251 -h a
l .n 6545 = s '? j 1 JUDGE BLOCil: The tests were not done under f 2 normal operating conditions? k 3 WITNESS CHEN: Yes, but I think they are =t 3p over and above the normal operating conditions. So 4 i 5 our concerns about the bolts would have been ~k ) o resolved by the fact that they were subjected to a much i i 7 higher load and did not fail, and furthermore, that 3 the allowables for the bolts for normal operatino conditions ? ensures that there will be a low yield. In fact, I 10 think they are.75 times the yield stress. i ' il MR. MIZUt!O: Chairman Bloch, if I could l ,j - 12 just ask you a few questions. )[ '3, JUDGE BLOCII: Please.
- l
- f BY MR. t1IZUNO:
i 3 15 Dr. Chen, are the allowables under the normal 10 16 conditions lower than the allowables under the LOCA . f, ' 17 conditions? ip;;p> la g A. (WITNESS CIIEM) Yes. Tj i 10 G If you tested the entire insert to failure 20 and you showed that it met the recuired factor of 21 safety, then of necessity, it had to go and show that the V 22 allowables for the normal and upset conditions were also '3 met? 'Ex = {g{( 24 A (WITNESS CIIEN) Yes. 25 JUDGE BLOCII: Of course, somewhat leading NI -1 4^W -!? ~. .jy,' L t Li:. 9 t. J
.m 6546 "1 ~ i l IT b redirect. But as I understand it, that is not necessarily .y-true. Because we are talking in the one case about a 2 ~ 3 one-time stress and in the other about a cyclical h s tress ; isn't that correct? .E. MITNESS CllEN: Yes. b 7; MR. SCINTO: Mr. Chairman, I thought the probic:- and the confusion was the point that at the test, with the LOCA condition, with the one-time stress there was deformation. The argument was that deformation was nonelastic behavior; therefore, the test did not 't j' l demonstrate elastic behavior under normal usage. i 8 f, i .1 4-i ? 4 - I _.4. ~ ?[ 1: 't ll '1 l r E[ I $I) i 'n, .O DF ~ 2 ,o r .6 b + ,q 7d 7 e n l t c' n a-d 16 U ? i g. J.;' 1 ?,E 4-
F~ 6547 T' n st 1 i JUDGE BLOCH: That is not what we are discussing i. 'y, - I now. I - l MR. SCINTO: I want to add that the witness f testified tha't'under normal loads those bolts act .i i elastically, because of the nature of the load. di r JUDGE BLOCH: Providing you stay below w
- r-
), l . yield right? You can go above yield -- i AR. SCINTO: I don't want to testify. I think the witness has indicated that. j. JUDGE BLOCH: Under normal loads you can' t go II l above yield; is that right? WITNESS CHEN: That's right. 5} '=b JUDGE BLOCH: Under a one-time load you can. jkV _ WITNESS CHEN: Yes.
- A V-'"
JUDGE BLCCH: We did a test which assured im i ~~ t i ourselves that we meet the accident conditions, one_ 17 time loads? ..dk WITNESS CHEN: Yes. e de i J h [j.c -j I JUDGE BLOCH: How did you extrapolate down from 47 . l
- i,
j those tests to assure yourself that the normal loads were "' w [? met? y ) WITNESS CHEN: That is not how I did that. N 23 3-JUDGE BLOCH: How did you do that?
- 4 gi WITNESS CHEN
By looking at the allowables Ws,. ,b e hs for the normal and upset conditions. [f ip ':f ll e d
. x t. . a,. N t. 4i = Manual of i STEEL CONSTRUCTEON b EIGliTil EI;ITION i 7 M I a r MCr CQn n.9titulC Of rStrel (.'tinstruction, fnc. ^ q .gg,, ? 3e m
~ s. ns.. AISC Specification (Effective 1!!!/78) 3 1.3.5 Mad 1.4.1.2 Unidentified a Proper provision shall be made for stresses caused by wind, both during for parts of minor importa erection and after completion of the building. physical properties of the st of the structure. !i 1.3.6 Other Forces 1.4.2 Other Metals Structures in localities subject to earthquakes. hurricanes and other ex-traordinary conditions shall be designed with due regard for such conditions. Cast steel shall conform edition: j j 1.3.7 Minimum Loads Afild-to. Medium. 1 In the absence of any applicable building code requirements, the loads re-Applications. ferred to in Sects. 1.3.1,1.3.2.1.3.5, and 1.3.6 shall be not less than those recom-High-Strength St. mended in the American National Standards Institute Building Code Require-Grade 80-50 1 i ments for Afinimum Design Loads in Buildings and Other Structures. ANSI Steel forgings shall cor A58.1. latest edition. edition: i Steel Forgsnes Cr AS W A668 SECTION 1.4 MATERIAL Certified test reports l 1.4.1 Structural Steel with the standards. l i 1.4.1.1 Material conforming to one of the following standard specifications 1.4.3 Rivets (latest date of issue) is approved for use under this Specification: Steel rivets shall conic StructuralSteel. ASTM A36 edition: Welded and Seamless Steel Pipe. ASTM A53. Grade B I High-Strength Low-AlloyStructuralSteel. ASTM A242 9 teel 5 High-Strength Low-Alloy Structural Afanganese Vanadium Steel. Manufacturer's certifica ASTM A441 with the standard. I Cold-Formed Welded and Seamless Carbon Steel Structural Tubing in Rounds and Shapes. ASTM A500 1.4.4 Bolts Hot Formed Welded and Seamless Carbon Steel Structural Tubing, ASTM A501 Steel bolts shall conforn-High-Yield Strength Quenched and Tempered Alloy Steel Plate, edition: Suitablefor Welding, ASTM A514 Low-Carbon Steei Structural Steel with 42,000 psi Minimum Yield Point. ASTM A529 Fasteners. AS Hot-Rolled Carbon Steel Sheets and Strip, Structural Quality, ASTM High Strength Bot A570, Grades D and E Nuts and Plai l High-Strength Low-Alloy Columbium-Vanadium Steels of Structural Quenched and Ter Quality, ASTM A572 Quenched and Ten a f. High-Strength Low Alloy Structural Steel with 50,000 psi Minimum gg3y g499 Yield Point to 4 in. Thick. ASTM A588 ? Steel Sheet and Strip, Hot-Rolled and Cold-Rolled, High-Strength. In connections, A449 bolt Low-Alloy,withimproved Corrosion Resistance, ASTM A606 quiring bolt diameters greates Steel Sheet and Strip, Hot-Rolled and Cold-Rolled. High-Strength, for high-strength anchor bol ) Low-Alloy, Columbium and/or Vanadium, ASTM A607 Manufacturer's certifica- ? ) Hot-Formed Welded and Seamten High-Strength Low-Alloy Struc-with the standards. l 4 turalTubing, ASTM A618 1.4.5 Filler Metal and F1-q Certified mill test reports or certifimi reports of tests made by the fabrica. tor or a testing laboratory in accordance with ASTM A6 or A568, as applicable. Welding electrodes and i and the governing specification shall constitute sufficient evidence of conformity cations of the American We' with one of the above ASTM standards. Additionally, the fabricator shall,if requeste,d, provide an affidavit stating that the structural steel furnished meets
- ^gY,
d the requirements of the grade specified. Commentary Sect.1.4. M l. J
Structural Steri for Buildings. s I1 1.4.1.2 Unidentified steel,if free from surface imperfections. :nay be used sses caused bv wind. both durin: for parts of minor importance. or for unimportant details, where the precise physical properties of the steel and its weldability would not affect the strength of the structure. quakes. hurricanes and other ex-1.4.2 Other hietals due regard for such conditions. Cast steel shall conform to one of the following standard specifications. latest edition: Afild-to-Afedium-Streneth Carbon-Steel Castings for Generai code requirements. the loads re. Applicatiors. ASTA1 A27. Grade 65-35 ,all be not less than those recom-High-Streneth Steel Castines for Structurai Purposes. ASTH 1 A148. 1nstitute Building L od. Require-Grade 60-50 nes and Other Structures. ANSI Steel forgings shall conform to the following standard specification, latest edition: Steel Foreings Carbon una Ahoy for GeneraiInaustriai Use. ASTH 1 A668 Certified tes; reports shail constitute surficient evidence of contormity with the stanciaras. - following standard specifications this Specification: 1.1.3 Rivets Steel rivets shall i onrorm to the tollowing st 'ndara specification. !atest
- Th1 A53. Grade B edition
Steel. ASTH 1 A242 Strei Structurai Rivets. ASTH 1 A502 -l Afanganese \\%nadium Steel. N1anutacturer s certihcanon snali constitute suf ficient evidence of conformity Carbon Steel Structurai Tubine "i'h'h''"d*' .500 'arbon Steel Structural Tubine, 1.4.4 Boits Steei bolts shall conform to one or the followmg standard specifications. latest -i Tempered Allor Steel Plate. edition: 14 mum Yieid Point. ASTH 1 A529 Loufarbon Steei Extanaily and Internally Threaaed Standara I g,,g h,,"n e t ? Itrip, Structurai Quaist v. ASTH 1 i s to tructural Steei Joints. Includine Suitable -- Vanaaium Steels of Structural "5" ""'"'"'" * * " " * ' ^ ^ Quenched and Tempered Steel Bolts and Studs. ASTH 1 A449 " Steel tvith 50.000 psi Afinimum Quenc and Temped AUoy weiBohs fodructuralmeeMoints. IA588 ASTA1 A490 id Cold-Rolled. High-Streneth, in connections. A449 holts may be used only in bearing-type connections re-eion Resistance. ASTH 1 A606 quiring bolt diameters greater than 1 % inches. A449 bolt material is also acceptable yid Cold-Rolled. High-Streneth. for high-strength anchor bolts and threaded rods of any diameter. anadium. ASTN1 A607 N1anufacturer's certification shall constitute suf ficient evidence of conformity -ich-Strength Low-Alloy Struc-with the standards. -ts of tests made by the fabrica_ l.4.5 Filler hietal and Flux for Welding
- STAI A6 or A568. as applicable.
Welding electrodes and fluxes shall conform to one of the following specifi-ufficient evidence of confortaity cations of the American Welding Society, latest adoption. as appropriate:* tionally, the fabricator shall. if
- structural steel furnished meets Approval or these welding electrode specifications in given without regard to weld metal natch toughnens requirements, which are generally not necessary for budding construction. See 5
Commentary Sect.1.4.
~ I $. 24 AISCSpecificatiors (Effective 11/1/78) 1.5.2.2 Design for rivets, bolts. and threaded parts subject to fatigue loading l.5.3 Welds shall be in accordance with Appendix B. Sect. B3. Exceptas modified by the prc to meet the stress requirements F' TABLE 1.5.2.1 ALLOWABLE STRESS ON FASTENERS. KSI AMVM Allowable Sheard (F ) l Type of Weld and Stressa e Abw. Friction. type Connections" able Bearmg. Complete I Oversized type i Tension normal to Samea Description of Fasteners
- Tension, Standard and Short.
long-Connec. 'N'C''' 4pg size slotted slotted tions. Compression normal to Samea Holes Holes
- Hole, effective area I
17.5' A502. Grade 1. hot. driven rivets 23.0" A502. Grades 2 and 3. hot. driven I Tension ar comoresuon Samea parallel to axis of weld p O' [ A307 bolts 20.0* 10.0kr Shear on effective area 0.30 j Threaded parts meeting the
- 'ci"py x
requirements of Sects.1.4.1 i metan yield s: and 1.4.4.and A449 bolts meeting the requirements of Partial.Fe Sect.1.4.4. when threads are Compression normal to Samea not excluded from shear planes 0.33F., *d 0.17F.,h effective area Threaded parts meeting the requirements of Sects.1.4.1 and 1.4.4.and A449 bolts p,,",$,g"g',',*,","oy','$on I'* Samea meeting the requirements of r parallel to axis of 30 Sect.1.4.4. when threads are excluded from shear planes 0.33F.,3 0.22F, h except e A325 bolts. when threads are not fji[, excluded from shear planes 44.0d 17.5 15.0 12.5 21 # A325 bolts,when threads are [,$',8sior. normal to effective 0 excluded from shear planes 44.0d 17.5 15.0 12.5 30.0' except l m gl,s A490 bolts.when threads are not excluded from shear planes 54.0d 22.0 19.0 16.0 28.0f A490 bolts, when threads are excluded from shear planes 54.0d 22.0 19.0 16.0 40.0f Shear on effective area 0.30 strenrt a Static loading only. e#'a s mt b Threads permitted in shear planes. yield st,
- The tensile capacity of the threaded portion of an upset rod. based upon the cross-sectional l Tension or compression Samea area at its major thread diameter. As.shall be larger than the nominal body area of the parallel to axis of weld' rod before upsetting times 0.60F,.
Ph d For A325 and A490 bolta subject to tensile fatigue loading.see Appendix B. Sect. B3. 0.30 Shear parallel to fayin$ sur. faces (on effective ares strenet
- When specified by the designer, the allowable shear stress. F., for friction. type connections
'8ceD5 having special faying surface conditions may be increased to the applicable value given Qjij3,j k in Appendia E. r When bearing-type connections used to splice tension members have a fastener pattern a For definition of effective area. see Si wheee length, measured parcllel to the line of force, exceeds 50 inches, tabulated velues 6 for " matching" weld metal. see Table shall be reduced by 20 percent.
- Weld metal one strength level strong-s See Sect.1.5A d See Sect.1.10.8 for a limitation on un b See Appendix A. Table 2. for values for specific ASTM steel specifications.
. Fillet welds and partial. penetration r members,such as flange.to-web cono i 8 For limitations on use of oversized and slotted holes, see Sect.1.23.4. or compressive stress m these elemes k
6 m StructuralSteelforBuildings e 5 25 ied parts subject to fatigue luadine B3. 1.5.3 Welds Except as modified by the provisions of Sect.1.7, welds shall be proportioned to meet the stress requirements given in Table 1.5.3. TABLE 1.5.3 AI.L.OWABLE STRESS ON WELDS Allowable Shear
- iF )
Required Weld Type of Weld and Stress
- Allowable Stress Strength Levet**
ction type Connections" Bearing-Complete-Penetration Groove Welds Oversized type Tension normal to Same as base metal " Matching" weld metal must tard and Short-Long-Connec-effective area be used. e slotted slotted tions' Compression normal to Same as base metal es Holes Holes effecuve area 17.5f Tension or comoression Same as base metal Weld metal with a strength parallel to axis et weid level equal to or less than 22.0' "matchmg ' weld metal may I go.ou Shear on et'fective area 0.30 x nominal tensile be usei l strength of weid metal (ksis. j except shear stress on base metal shali not exceed 0.40 x l yield stress or base metal Parual-Penetration Groove Weldsd l Comoression normal to Same as base metal i 0.17f,h effective area ?. - 1 I Tension or compression Same as base metal parallel to axis os weld' l l' Shear parallel to axis of 0.30 x nominal tensile Weld metal with a strength weld strength of weld metal (kso. levet equas to or less than except shear stress on base "matchmg
- weld metal may l
0 22f=, metal shall not exceed 0.40 x be used. i 15.0 12.5 21.0f yield stress of base metal Tension normal to effective 0.30 x nominal tensile area strength of weld metaliksi>. 15.0 12.5 10.0, except tensile stress on base metal shall not exceed 0.60 x 1 19.0 16.0 28.0f yield stress et base metal Fillet Welds 19.0 16.0 40.0' Shear on effective area 0.30 x nominal tensile strength of weld metal (ksi). except shear stress on base Weld metal with a strength metal shalinot erceed 0.40 x level equal to or less than yield stress of base metal "matchmg weld metal may be used' t rod. based upon the cross-secuonal Tension or compression Same as base metal r than the nominal body area of the parallel to axis ot' weld
- Plug and Slot Welds I
ading.see Appendix B. Sect. B3. t ress. E,. for incuan-type connecuons Shear parallel to faying sur. 0.30 x nommal tensile Weld metal with a strength faces ton e fecuve areas strength of weid metal eksis, level equal to or less than ~ reased to the applicable value given except shear stress on base " matching ' weld metal may metal shall not exceed 0.40 x tn used. in members have a fastener pattern exceeds 50 inches, tabulated values
- For defm' ition of effective area. see Sect.1.14.6.
6 for " matching" weld ruetal. see Table 4.1.1. AWS D1.1-77.
- Weld metal one strength level stronger than " matching" weld metal will be permitted.
.y steei specifications. d See Sect.1.10.8 for a limitation on use of partial-penetration groove welded joints. sao Sect.1.23.4.
- Fillet welds and partial. penetration groove welds joinm' g the component elements of built-up members, such as slange-to-web connections may be designed without re or compressive stress m these elements parallel to the axis of the welds. gard to the tensite l !lp 1
I 31, . L y Q __ l
e IN I L l s = s - ? L J.n.-,1, UNntu STATES NUCLEAR REGULATORY COMMISSION ~ IN THE MA1 Ar.R OF: DOGGT NO: INDEPENDENT ASSESSMENT PROGRAM - / COMANCHE PEAK STEAM ELECTRIC STATION LOCATION: BETHESDA, MARYLAND PAGES: 1 - lig DATE. THURSDAY, DECEMBER 20,'1984 j ~ ace-FEDERAL REFoarse, Ixc. Onid 1F.ercriers 444 'E:-h Seitc!Stree: Wash::. :::.. b.C. 20001 '202'; I47-37C0 NAT:0.'A C ECC'.E ACE [ = 6
( i .w -g h; 1 -i' 14 04 80 tb 1 separate editions of the AISC manual were employed by WWeb [(;j$ l ( f-2 different design organizations at Comanche Peak. General ~x r, 3 Note 9 to our pipe support checklist on Phase 3 summarized C.; l'S 4 the examples which CYGNA found during our review of use of
- y....
5 the 7 th and 8th editions of the AISC manual. fun ~ i:.'t j Q;,: 6 We found no design impact and in fact TUGCO later _ p{p 7 issued a DCA, a design change authorization, which changed a 8 their pipe support design specification MS46A to adopt both (! 9 of those editions. i cm ,w.- 10 Item 10, cable tray dampi ng valuer This was .< F.:is <; i% 11 particularly born out of the hearings which we participated
- ur-e 12 in.
It may or may not be one of the issues on the S E.~ I 13 Walsh/Doyle list. The discussion at that point in time 14 , centered around the use of welded structure damoing values i 1-15 from Reg. Guide 161 versus bolted. g 16 CYGNA still stands behind its position that we f. 17 provided in response to Walsh Question No. 5 in our prefiled 1 18 -testimony where we feel that the use of damping values for ] e 19 bolted structures for cable trays as a system was perfectly ig 20 appropriate. [ 21 I have also down here some reference to the Phase 22 4 review. That's because cable tray supports and conduit g_;
- .D s
23 supports are specifically part of the Phase 4 review so ( .e.
- s..e..
E ' b h(j 24 there will be some further documentation on their position 25 in that report. []:9 { 4 f2yj-g .7 u.q s e
- - J ~
UNITED STATES OF AMERICA K4ED NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BUkRDdli 18 A!0: In the Matter of I r rn - I / TEXAS UTILITIES GENERATING l Docket Nos. 50-445-1A COMPANY, et al.- l and 50-446-y I (Comanche Peak Steam Electric Station l Station, Units 1 and 2) l CASE'S ANSWER TO APPLICANTS' STATEMENT OF MATERIAL FACTS AS TO WHICH THERE IS NO GENUINE ISSUE REGARDING THE EFFECTS OF GAPS ON STRUCTURAL BEHAVIOR UNDER SEISMIC LOADING CONDITIONS' in the form of AFFIDAVIT OF CASE WITNESS MARK WALSH l'. Applicants state: "All bolts in multiple bolt, bearing-type connections will react imposed shear loads within at most the distance of the bolt hole tolerances. (Iotti, Finneran' Affidavit at 8 f1f.)" I disagree with this statement (and with some of the statements in the Affidavit). At the point all bolts begin to react the shear loads (that'is, when the last bolt will have received a 1 lb. shear load), the first bolt that has reacted the shear load may have the shear load of 1,000 lbs., and this may have exceeded-the~ allowable shear capacity of the bolt. This is assuming that the first bolt that' reacted the shear load has not failed when the last bolt begins to resist the shear load. - [1/ I believe that the actual citation should be at 4-5. 1
One of Applicants' primary arguments which is contained in the back-up Iotti/Finneran Affidavit is regarding the definition of " oversized" bolt holes. This is addressed in detail in answer 2. following. One example of statements with which I disagree which is made in the Affidavit of Dr. Iotti and Mr. Finneran is found on page 5 of the Affidavit (last paragraph, continuing on page 6), wherein they cite " Structural Design Guide to AISC Specifications for Buildings," by Paul F. Rice and Edward S. Hoffman (Attachment B to Affidavit). Although the one page (268) from that document which Applicants have attached appears to be accurate /2,/, their discussion and the portion cited (which is out of context) are very misleading. (See Attachment A, pages 264 through 271 of the Rice /Hoffman text.) In the portion attached by Applicants, Messrs. Rice and Hoffman are only talking about connections that receive static loads (i.e., loads that do not change direction) because the yielding stress criteria is not applicable in friction-type connections, as will be discussed later. The connections referenced by Messrs. Rice and Hoffman specifically exclude the supports that have dynamic loads (as will be shown below) such as most of those supports at CPSES. In addition, the inelastic deformation in bearing-type connections is recognized by the AISC Code at 1.5.2.2, where the allowable bearing stress is 1.35 Fy. Therefore, the Applicants' statements are lacking reference to the fj2/- It should be noted that I have not reviewed the entirc text of the other reference cited by Applicants, " Plastic Design of Steel Frames," and cannot state whether or not it is taken out of context. 2 i J
l ) i specific amount of inelastic deformation allowed by the ATSC Code for a non-dynamically loaded structure. On pages 265-266 of the Rice /Hoffman text, following a discussion lL of'AISC and ASTM specifications, it is stated (the numbers Messrs. Rice f and Hoffman have placed in parentheses refer to sections from AISC or i ASTM): "The use of ordinary (A307) bolts is limited by a number of 4 Specification requirements. The allowable stresses are low: tension Ft = 20 kai on the threaded area; and shear Fv - 10 kai i (1.5.2.1). The slip before full bearing is achieved on a group of ordinary bolts effectively rules out the sharing of stress in a mixed connection. Holes are to be taken as 1/16 in. larger than the nominal diameter (1.23.4), and the ordinary bolt does not expand to fill.out the hole like a driven rivet nor can it be used for dependable friction. Stress sharing may not be assumed c between ordinary bolts and rivets or welds (1.15.10; 1.15.11).- In addition, low-strength bolts are not permitted in important field connections including... connections subject to vibration. impact, or stress reversal (1.15.12). . " (Emphases added.) As discussed above, according to Messrs. Rice and Hoffman ( Applicants' own chosen authority), A307 bolts are not permitted in [ connections subject to vibration, such as'those at Comanche Peak. i 1 Applicants have admitted that the-connections at Comanche Peak are subject to vibration. Applicants' witness Finneran' stated, in regard L to the support which Jack Doyle and I noticed that had failed during hydrotesting (Tr. 4793/15-4794/4): "0: (By Mr. Reynolds) Would you render an opinion on why the paint that Mr. Doyle. talked about may have flaked during the flow of fluid through the pipe? "BY WITNESS FIN?ERAN: "A. I would say that possibly vibration may have been a cause for paint coming off of the deformed area during flow of fluid during the pipe; one possible cause. "Q. Yes. So what you are saying is that it could have been that when the deformation was caused during construction, the paint cracked but remained on, and then when vibration occurred due to hydrostatic flow, the paint chipped off? 3
"BY WITNESS FINNERAN: "A. It's a possibility. I couldn't say if that's exactly what happened." (Emphases added.) And at Tr. 5002/24-5003/9, Mr. Finneran further testified: "Q: (By Mr. Walsh) Is vibration a common occurrence at Comanche Peak? "BY WITNESS FINNERAN: "A. I think all piping systems that have fluids in them or flowing to them are possibly subject to some vibration. "Q: How about other pipes, main steam? Will they have vibrating effects? "BY WITNESS FINNERAN: "A. Quite possibly there will be vibration in the main steam piping." (Emphases added.) The ASME Code requires the Applicants to minimize vibration where it states: "NF-3112.2 Design Mechanical Loads. . The requirements of (a), (b), and (c) below shall apply. "... (c) Component supports shall be designed to minimize vibration." In addition, according to Messrs. Rico and Hoffman (Applicants' own chosen authority), Applicants are also barred from using A307 bolts because of stress reversal. As indicated above, Messrs. Rice and Hoffman cited the AISC code (to which the Applicants are committed through Specification MS-46A), Section 1.15.12, which states, in part: " Field Connections " Rivets, high strength bolts or welds shall be used for the following connections: ... Connections for supports of running machinery, or of other live loads which produce impact or reversal of stress. "In all other cases field connections may be made with A307 bolts." (Emphases added.) 4
CASE requested, through discovery on the issue of Applicants' Motion for Summary Disposition regarding generic stiffnesses, the drawings which the Applicants used in their Motion. Of the 60 supports which the Applicants provided (I count 59, but this is immaterial to this point), 52 had reversible loads which is a reversal of stress on the supporting connection. (See Attachment B, the referenced 59 drawings.) On the drawings, the reversal of loads is shown in the block listing the loads and the direction of the load is indicated as + or . Of the 7 supports which do not contain reversible loads (CT ' 013-006-S22S, CT-1-013-004-S32S, MS-1-001-002-C72S, MS-1-01-001-C72S, CT-1-013-002-C42S, CC-2-011-719-A53R, CT-1-013-011-S22R), 5 are spring j cans and 2 are rigid-type supports. Based on this random dample, l 88% of these supports require high strength bolts due to the requirement of a reversal of stress (load) f3/, according to the AISC Code (to which the Applicants are committed in design specification MS-46A). It is also obvious from the preceding discussions that Applicants are in violation of ANSI N45.2.11, 3. DESIGN INPUT REQUIREMENTS, 3.2 Requirements, which states, in part: "The design input requirements should include the following where applicable: "(9) Mechanical requirements such as vibration, stress, shock and reaction forces." (Emphases added.) /3/ Stress is equal to the load divided by the cross-sectional area of the item under consideration. S
1 ~ a ATTACRIENT A n Structural DesignGuide toAISC Specifications forBuildings i PaulE Rice Edward S.Hoffman f I ( VAN NOSTRAND REINHOLD COMPANY _ _ = 1,,. =. = y _ _ i
l l t This book is intended Van Nostrand Reinhold Company Regional Offices: tine designs with the s' } New York Cincinnati Atlanta Dattas San Francsseo steel and joist const,ruct Van Nostrand Reinhold Company International Offices: Each new AISC Spe. London Toronto Melbourne language for safe stru, plastic design, each n Copyright c 1976 by Litton Educational Publishing. Inc. economy of materiali Library of Congress Catalog Card Number: 75 40491 design calculations. ISBN: 0 442 26908-8 bcreasing comp All rights reserved. No part of this work coves.d by the copyright hereon may for design. Computer be reproduced or used in any form or by any means-graphic, electronic, or howcVer, and compute. mechanical, includmg photocopying, recordmg. taping. or information storage This Guide does not and retrieval systems-without permission of the publisher. the hianual of Steel ( Manufactured in the United States of America Inentary, shows how t. ggan. It converts s Published by Van Nostrand Reinhold Company to direct design. It } 456 West 33rd Street,New York,N.Y.10001 design. Published simultaneously in Canada by Van Nostrand Reinhold Ltd. Tables 31 and 3 2 15141312111098765432 AISC equations for a beam section, or solu about three minutes. Specification requir Library of Congress Cataloging in Publication Data frames, composite cc 1921-Rice, Paul F Structuraldesign guide to AISC specifications for based on structural s buildags. cussed to aid the s specifications. Includes bibliographical references and index.
- 1. Structuraldesign-Handbooks, manuals,etc.
It is assumed that u
- 2. Building-Contracts and specifications.
I. floff. as well as the terms t man. Edward S.1920- joint author. IL Title. requirements in the < TA558.3.R52 690 75 40491 ISBN O-442-26904-8
it is not intended in 1 in the A1SC Handboul. presentation of a wide search in this area. (Se-explanations and illustr overlooked or troublesc able interpretations to r extension of such inter; been desirable to extent plate connections were CONNECTIONS e le re; Code. During theinter revised to agree with thi Rivets, Pins, and Bol Rivets and Pins. The t-years ago and were in been directed toward there has been little ch the requirements for, a General in little need for inter The latest AISC Specifications permit a wide variety of connections. The basic require. available under ASTM ment, appropriate with the sophisticated wmbinations of different types of steel to be are given in Table 1.5 connected, different design requirements of connections, and different means of connec. (1.14.2); computeti as tions, is a performance requirement consistent with the overall development of the diameter of the rivet h Specifications. This requirement states simply "..that the design of connections be The use of pin com consistent with the assumptions as to the type of construction...."(1.2). Each of the modern steel building, detailed requirements for the design of connections simply builds upon this basic require. ing special design. Tht ment. By implicitly or explicitly requiring that the design of a particular type of connec. unchanged from previo tion be consistent with the design assumptions as to the type and amount of force to be Perhaps the rnost ust transmitted, and rotation capacity (or ricidity) consistent with the rotation assumed attuations or addition necessary to develop the connection forces, the basic performance requirement is constmetion, beargg-completed (1.2). (1.15.10). If used in i The Specifications explicitly recocmze melastic behavior in connections of members strengthening existing designed as clastic: " virtually unchanged" angles at the joints in rigid frames, "non. place loads,and the ne elastic" deformation of parts of connections in Type 2 and 3 construction, and Bolts. Bolts may be c " inelastic rotation" for wind connections with Type 2 construction (1.2). Elastic be. A325, for f, = 40 ksi, havior in the connections of members under plastic design is implicitly recognized (2.1). (A307) bolts are usar Scope may be designed for ei The use of ordinary For the purposes of this Chapter, connections are most conveniently considered as The allowable stresses classified on two bases; (1) materials used (rivets, bolts, pins, or welds), and (2) the as. 10 ksi (1.5.2.1). The sumed behavior of the connection (design requirements: rigid, semi. rigid, or plastic for fectively rules out the moment; shear transmission only; tensile or compressive force only; or combinations). k in. larger than the n In addition to forming joints between two or more steel members or parts of members, fill out the hole like a connections are required to elements composed of other structural materials. For com. ing may not be assun posite action with concrete elements not bonded by encasement, shear connections are in addition, love.stren required (1.11.1). Shear connections may utilize specially designed shear connectors column splices in all or standard welded stud connector. (1.11.4; 1.4.6). For connection of steel column where width / height < bases to transmit any direct tension or bear, anchor bolts are required (1.22). 264
~ m.- -m 1 I CONNECTIONS 265 j It is not intended in this chapter to duplicate the design aids, detail data,and examples in the AISC //andbook, Part 4, Connections. Equally, space limitations do not permit presentation of a wide range of examples to !!1ustrate even the recently published re. search in this area. (See " Selected References".) Rather, the purpose here is hmited to explanations and illustrations of all applicable Specification requirements that might be overlooked or troublesome in routine work. This aim will include indication of reason. able interpretations to resolve apparent conflicts or ambiguities in the Specifications,and extension of such interpretations where the Specifications seem to have omissions, it has been desirable to extend this aim somewhat in that design aids for bearing plate and base plate connections were included as well as an extension of concrete bearing connection l design to cover an apparent gap between the AISC Specifications and the ACI Building .r.3 Code. During the interim between preparation and publication, AISC specifications were revised to agree with the latest ACI BuildingCode. Rivets, Pins, and Bolts i Rirets and Pins. The requirements for the use of rivets and pins were established many years ago and were in many AISC Specifications. Since most of the late research has been directed toward welded, and more recently high. strength bolted connections, there has been little change in the Specifications for the use of rivets. Familiarity with the requirements for, and a sharply reduced use of, rivets in building construction results in little need for interpretations of these Specifications. Rivets of Grades I and 2 are i available under ASTM A502 (1.4.3). Allowable stresses (for tdnsion and bearing only) The basic require. are given in Table 1.5.2.1 (1.5.2.1). Net sections for tension members must be used types of steel to be (l.14.2); computed as prescribed (l.14.3); and allowance of ;Ig in. made plus the nominal it means of connec. diameter of the rivet holes (1.14.5). evelopment of the e l The use of pin connections, originally popular in truss construction, has declined in . of. connections be modern steel building, and is usually encountered only for very special situations requir. r (1.2). Each of the ing speciai design. The general requirements for the use of pins are brief and essentia!!y in this basic require, unchanged from previous Specifications (1.14.6). alar type of connec. Perhaps the most used application of these Specification requirements today will be in tount of force to be alterations or additions to existing buildings in which rivets or pins were used. For new e rotation assumed construction, bearing. type connections can not be assumed to share stress with welds 1ce requirement is If used in combination, the welds must be designed for the entire stress. In (1.15.10). strengthening existing construction, bearing connections can be assumed to carry the in. tctions of members l place loads, and the new welds designed only for the additional stress (1.15.10). rigid frames, "non, construction, and Bolts. Bolts may be classified by strength as(1) low, A307,for F, = 20 ksi;and(2) high, i (l.2). Elastic be. A325, for F, = 40 ksi, and A490, for F, = 54 ksi... (1.5.2.1). The ordmary low strength ly recognized (2.1). (A307) bolts are usable only in bearing connections (1.5.2.2). The high strength bolts may be designed for either bearing or friction connections (1.5.2.1). The use of ordinary ( A307) boltsislimited by a number of Specification requirements. ntly considered as The allowable stresses are low: tension F, = 20 ksi on the threaded area; and shear F, = ds), and (2) the as. 10 ksi (1.5.2.1). The slip before full bearing is achieved on a group of ordinary bolts ef. fectively rules out the sharing of stress in a mixed connection. Ifoles are to be taken as rigid, or plastic for { ig in. larger than the nominal diameter (1.23.4), and the ordinary bolt does not expand to
- or combinations).
l ut the hole like a driven rivet nor can it be used for dependable friction. Stress shar. parts of members fill saterials. For com$ [ ing may not be assumed between ordinary bolts and rivets or welds (1,15.10; 1.15.1I). rar conacetions are l in addition, low. strength bolts are not permitted in important field connections including e lumn splices in all buildings with 1/> 200 ft., and where width / height < 0.25; also d shear connectors I in of steel column where width / height < 0.40, for ll> 100 ft.; beam. column or column. bracing connections ( g,22), l. Af j l {
STRUCTURAL DESIGN CUIDE TO AISC SPECIFICATIONS FOR EUILDINGS 266 innec-and the member connec: where # > 125 ft.; frames carrying cranes with more than five-ton capacity;and to accommodate the nec tions subject to vibration, impact, or stress reversal (1.15.12); nor fcr flange.to wsb nor and the selection of the ci cove plate-to-Gange connections of built-up girders (1.10.4). High strength bolts (1.16.1) and welds are considered essentially equivalent as connec-As previously noted, v tions, and, for friction type joints assembled prior to the welding, the high-strength bolts quired as a mixed conr may be assumed to share stress with welds in a mixed connection (1.15.10) or with rivets required prior to the we' (1.15.11). Gross sections may be used for the design of compression members (1.14.2), also important, though : and for the flanges of both built.up and rolled-shape girders provided the area of holes is generated in the opera' strained, leave correspo. equal to or less than fifteen percent of gross Gange area (1.10.1). For tension members net section area is the basis of design (1.14.2). In friction-type joints resisting direct local inelastic yieldmg, I tension, the shear stress permitted with high-strength bolts must be reduced (1.6.3). warping and lamellar tes caud n a sPecMe Slotted Holes for Bolted Shear Connections. The use of short slotted holes is permitted av id warping. Even af under 1974 AISC Specincation for " Structural Joints Using ASTM A-325 or A-490 to Bolts," Section 3. subject to the approval of the designer. They can be used in either E' d ss ef friction-type or bearing. type connections, provided a washer is installed over the hole. service is n 1 Provided us The normal hole size for a j" d bolt is Q'.whereas a short. slotted hole is M" deep by The use of a proper i 1" long (or d" longer in the horizontal dimension). While the Specifications state that
- ""I*
- P" the hole can be either vertical or horizontal, the authors suggest only the horizontal
"'* Y ** slotted method be used. End clip holes only would be slotted,not the holes in the con- *"*I"' nection beam or column. See sketch, a full scale view of the end clip holes and bolt re. should be given to the lationship for a typical d" thick web. E "' ** The advantages to this system are many, several of which are: the entire connection i 2,9 je," Ga. p 5/g,u%
- 1. Greater erection speed with less field
/ economical to specify a buming of misaligned holes. l on c.up t. 1. (Holea F
- 2. The use of one size clip angle with a Connect. ion Design set gauge will accommodate web thick.
f 3"x[- ness from d" to d" (j 9 } Classification. In addi1 $[ Bolts d 3 I connections, certain ai
- 3. The reduction in sizes of clips to fabdcate and stock should help i
.2 established. All connc< reduce costs. support not less than '
- 4. The speed of erection (and eh. ina-SE on centers i
m shear in flexural mem' i tion of mill web thickness tolerance mmb rs, ati at a11ow.a Shoet slotted holes layout problems) should help reduce cost. tor clip lS -shear connection (1.15.2). These minin1 Welds of light members such bers are required to. General. Full penetration groove. welds can be designed for full development, same the minimum six ki stress as the base metal (1.5.2.1), by selection of the specified matching electrode and 8""Ith of the memb-welding process (1.17.2). For all fillet, plug, and slot welds, and partial penetration truss m which the mi groove. welds, reduced permissible stresses upon the effective throat area (1.14.7) are I ad for open web ste specified (1.5.2.1). In no case may the stresses exceed that for the base metal,or if dif. design stress or half ferent,the weaker base metal (1.5.2.1). Joists; Examples.) Special Considerations. A number of minor special considerations arise in the specifica-As noted previously tion of welding. Generally, net sections are not a consideration except for plug and slot method or the design welds in which the gross area of the holes is deducted to check the fifteen percent maxi-the transmission of si mum allowed (1.10.1; 1.14.3). The Specifications require preheating for various con-bers connected and tr. ditions, including all work when the temperatures are below 32*F (1.23.6). Except for single and double. angle or similar minor members, welds are to be laid out to avoid ec- ,y centric axial force or such eccentricity must be considered in the design of the connection
- 3973,
l ~ .UBLCINIS CONNECTIONS 267 apacity;and connec-and the member connected (1.15.3). For the usual shear connection requiring flexibility or flange-to. web nor to accommodate the necessary simple.end rotations assumed, the locations of the welds and the selection of the connection elements must be coordinated (1.15.4). I ?quival:nt as connec. As previously noted, where welding at high. strength bolted friction type joints is re-e high-strength bolts quired as a mixed connection with shared stress, the final tightening of the bolts is 15.10) or with rivets required prior to the welding. The sequence of completing purely welded connections is n members (1.14.2), also important, though not explicitly covered by the Specifications (1.23.6). The heat d the area of holes is generated in the operations of welding creates intense shrinkage strains which, if re-or tension members strained, leave corresponding residual stresses (1.23.6). These stresses can be relieved by sints resisting direct local inelastic yielding, but where local inelastic yielding is also restrained or limited, duced (1.6.3). warping and lamellar tearing
- may result. For many welded assemblies, the simple pre.
cauti n I a sPecified sequence of welding may be employed to balance the strains and to d holes is permitted av id warping. Even after this precaution. certain complex assemblics may be expected 'M A.325 or A-490 an be used in either to retain adverse residual stresses. For cases where this condition is anticipated or sus-rd over the hole' Pected, stress relief by heating must be specified by the Engineer (1.23.6). (Note: this Ih:le is M" deep by service is n ! Provided unless it has been specified and will normally be an added cost.)
- ifications state that The use of a proper sequence to avoid creation of shrinkage stresses or to minimize same can als be specified in many connections where lamellar tearing might occur, only the horizontal the holes in the con-Particularly with thicker sections,where both the direction of the shrinkage is completely restrained and the resulting stress is normal to the surface of the section, consideration ip holes and bolt re-should be given to the welding sequence. If the condition can not be eliminated by a practicable sequence as a first choice for a solution,it may be possible to relieve the strains without developing large stresses by use of soft wire " cushions" or by revision of
.5/IS%ab the entire connection detail. At least for simple cases it should, of course, be more economical to specify a particular welding sequence. (See Examples this chapter.) Hsins t Connection Design I Classification. In addition to the general requirements previously cited for the design of l I connections, certain arbitrary minimum design requirements for connections have been l established. All connections for members carrying calculated stress must be " designed to ont:rs _; support not less than six kips" (except lacing, sag bars, and girts), presumably six kips ( shear in flexural members, six kips tension in ties, and six kips bearing in compression 1:s laym members, all at allowable stress levels (1.15.1). Eccentric connections of axially loaded members are to be designed to transmit the resulting moments as well as the axial force e ccnnech.on (1.15.2). These minimum requirements naturally become most significant in the design of light members such as axially loaded members in trusses. Connections for such mem-dIvelopment, same bets are required to meet an additional requirement that they transmit the design load or ching electrode and the minimum six kips, whichever is larger, and develop at least half of the effective partial penetration strength of the member (1.15.7). Note: joists are regarded as a special very limited. size it area (1.14.7) are truss in which the minimum connection capacity is simply specified as twice the design use metal, or if dif. I ad for open web steel joists (4.5); or for the longspan and deep longspan joists,as the design stress or half the allowable strength of the member (103.5b). (See Chapter 4: Joists; Examples.) ise in the specifica. As noted previously, connection types may be classified on the basis of the connection 8 i pt for plug and slot method or the design function. Broadly, connections may be described as flexil (for i fteen percent maxi-I the transmission of shear only,1.15A),or rigid (maintaining the angle between th nem-l ng for various con-bers connected and transmitting full moment capacity of the most flexible element at the .23.6). Except for d ef the ectio l l
j ku 268 STRUCTURAL DESIGN GulDE TO AISC SPECIFICATIONS FOR BUILDINGS [ joint as well as the shear,1.15.5), or semi-rigid (transmitting a pre determined fraction of T --- > the full moment capacity as a rigid joint and further loads in shear as a flexible joint with s 2 corresponding angle change to supply rotation for the additional loads,1.15.5; 1.2). Mexible Connections. " Flexible" connections are designed to transmit shear without /__ excceding allowable unit stresses on the connectors as a group or the connection as a whole. The use of an average capacity for each of several connector elements sharing the a Inelast total load is justified by allowing self-limiting localized stresses determined by an clastic joint analysis to exceed the yield point and create inelastic localized deformations of the plates ; connector materials, or by inelastic deformations of the connection elements (1.15.4). ginnin g tinuing-The simplest examples of localized deformation occur in the assembly of bearing type bolted connections where the cumulative tolerances permitted exist on (1) out of.round s2. s ).i. in the bolts,(2) oversize holes (gg"), and (3) center to-center location of the holes in the in all _ l different elements connected. The extreme degree of suchinelastic action occurs with a FIG *5 2 Self Li two-bolt bearing type connection where one bolt is loosely fitted and one is very tight. Until the material of the connected element surrounding the loaded bolt or the bo!t yields and deforms (+;1 "), the load is not shared and a 50 percent adjustment will be developed ,= 6 as the load increases. For larger (and thus more important) members, more bolts or rivets m will be required and the degree of adjustment required on each will be less. Lesser adjust. 8= ments are required for a long line of bolts or rivets intended to share stress equally. Even N @] if perfectly fitted, yielding and inelastic deformations occur, maximum at and beginning at the first loaded bolt or rivet, and decreasing to a minimum at the last. (See Figs.51 [ ] u~- l h-HFace of Supper t p A _t , is i i'. 'o I bt l +qI i , !M_. L I-f g j h} f m .a ) \\ l l.5 ( tuv A .o T EQ 1 o 9 2 FIG.5 3 Self Limi-y m I L -N-1 o Angles. g ~~~~ A)f y dt W u use of an average si I f e V/Dt until(6Tg*Obd): f/l6" served in one series bl e I (dg+6botl I"CP'" ""til fbl" IN
- I hmits correspondin b2 p
mately the same for FIG. 51 Self Limiting, Localized Deformations-Two Bolts. Figure 5-4 presents reported from thes "s" (see Fig. 5 3) and 5 2.) After this localized inelastic adjustment in the connectors for shear trans. mission, consider the inelastic adjustments that occur to reduce the " clastic theory" support. Coping t-and required angle moments. inelastic deformation in the connection elements, typically ang!:s, will occur and re. deep connections, duce the restraint which would transmit momant. The common double angle shear bear. For Type 2 const design drawings; al ing connection is extremely stiff longitudinally for the transmission of shear,'and it capacity for the se depends upon the minor inelastic bearing deformations around each fastener to equalize the shear stresses in the fasteners. The same double angle member is relatively flexible signed for one half and will twist to permit a relatively large angular rotation reducing moment transmission. (See Fig.5 3.)
- s. goment. Rotation Experience and tests confirm the practical assumptions of shear transfer only and the necting/ournal. AIS(
' J !!Y L,. l g. L e-.; - G.,
/ l s l l 1 Kult. DINGS CONNECTIONS 269 l ' determined fraction of 7i_ T as a flexib!c jomt witi. ids,1.15.5; 1.2). l $ 4 $ 4 h h; O Ol OO O l *ansmit shear without J l it the connection as a SoI8 lSsIS4l 2 't elements shanng the ermined by an elastic Inelastic defnrmations occur successively in raa d deformations of the = elates at each fastener and in the fasteners, be. l On elements (1.l$.4). Rinning and largest at the first loaded, and con. j 'mbly of bearing type tinuing until the elastic strains in the spaces og, j t on (1) out-of round becerre proportional tu equal stress s2' 83, and sg on of the holes in the in all fasteners, e action occurs with a md one is very tight. FIG. 5 2 Self Limiting Deformations-Axial Stress on Line of Separate Fasteners. Jolt or the bolt yields ent will be developed i, more bolts or rivets pFoce of support i te less. Lesser adjust. 13 q
- stress equally. Even d'
- -A
~ num at and beginning f-7 ~~} s e last- (See Figs. 51 [ 1 Q m -r s ( 0 o M 4 a.4L until := o $s d
- O ML
~ 4 24 F.1 1 F.1 = end morent l .( $=endrotatton l l l l s = space, tiottom flance )d g to race or the support FIG.5 3 Self l.imiting Deformation (Twist) in the Connection Elements ( lL) Two P i Angles. z'. use of an average shear stress per unit weld or separate fastener. The actual rotations ob. served in one series range from 0.84 to 0.97 times <bo, the " simple beam rotation." These i*Ib2 limits corresponding to moments ranging from three to sixteen percent were approxi-mately the same for a single end plate connector or the common double-angle connector.' S I'8-Figure 5 4 presents the usual device for an approximate analysis. An additional caution i reported from these tests is that the moment stiffness increases abruptly when the space )rs for shear trans. "s" (see Fig. 5 3) closes and the lower flange transmits compression to the face of the le " clastic theory" support. Coping the bottom flange where a quick analysis of the proportions of depth and required angle change show the usual clearance to be inadequate may be desirable for . will occur and re. [ deep connections. For Type 2 construction (flexible connections) all the reactions shou d be shown on the le-angle shear bear. a in of shear, and it design drawings; alternatively, only those exceeding one half the tabulated uniform load 'astener to equalize /, capacity for the sections used, together with a general note that connections shall be de. s relatively flexible B signed for one half the capacity unless otherwise noted, should be shown. ment transmission, h 8"Aloment Rotation Chara.teristics of Shear Connections," Kennedy. October,1969, 6, No. 4, Ensi-nsfer only and the neennsJournal, AtSC.
- k. n.~
. L
c e- { STRUCTURAL DESIGN GutDE TO AISC SPECIFICATIONS FOR BUILDINGS 270 'E E E.M.=wf 2/12 M>k 2 f I' Beamline 4=wf3Mf t gg y A Typical 12ance for 2 b.(JL) 7 connections - 0.844,io 0.97 +, f 5 E M4 Curves ,f-p @, Range
- g M=0 3
i c Simple ....: tg, g 5 Span v' , vd/24 EI 4*0 M di%4- / oEE M = wl Y12 M,= end moment in g e connection f, Fia.s e 6 Mc= center of span M p_ M, My= yield M and the " rigid" e pI 8. $=(2M f-Mf)/6EI y determined end r i b e = usi 2=M velop the yield r k 9 y I /24 practical cases wi CQ
- e( -
L required, the exe hinges only will f " 0 M, = My more rotation ca $ 7 (Mc= My pie ible Ni Masonry Beari l y 8 General. The A1 h_ o di
- I ',
on masonry and 7 o o.S
- 4, low allowable s',
regulations (1.5. i 1.ine@@: M = My Line@@:Me= My c block, and holl E FIG 5 4 Uniformly 1.oaded Elastic Beam Line-Rotations at Connection, terials. Values! i ti n is therefort _ Rigid Connections (Type 1 Construction). The AISC Specifications requirement is quite and brick laid i realistic: that rigid connections hold the original angles " virtually" unchanged (1.2). This of the inasonry requirement in clastic design is usually satisfied by connections designed to develop the ti naf associatk 1 f full section of the flexural member or the full moment at yielding of the more flexible AISC Specifica j r member connected. It will be noted from Fig. 5 4 that the rigid frame analysis (4 = 0 at fractions of tl y the allowable stress) may be satisfied by such a connection which would have a very smal; economy. 1; (I,! rotation at 0.66 to 0.60F,but would be capable of a significant rotation at yielding of codes. For be. =- y mend the use c g Il the flexural member (line A.B). A diagram similar to Fig. 5 4 but with point A representing an end moment, AI, = Np, and point C, a center span moment, AI, = Al,can be prepared for plastic design. Connec-Beam Bearing - p tions capable of achieving full collapse load (hinges at both ends and center of span) beam bearing - ] would be required to reach + = 0.54e, point B. The simpler concept of " plastic redesign," where only end hinges are required to form at the factored load,would require conne:tions with a somewhat less rotation capacity,along line A.B. See Fig. 5 5. Semi. Rigid Connections (Type.1 Construction). (See Fig. 5 4.) Ideally, the semi. rigid
- Atx mortars un
" supplement connections for Type 3 construction will behave clastically between the 4 = 0 ordinate g 1 v.
--_,_.yg, 7w CONNECTIONS 271 . DINGS Beatm Line @@@ \\@@'M.= M pb M P N .M =Mp e \\ .d 1 M = M, M e \\p,W .ons i f-N 17 &, i M M = M, c \\ O 2 sple \\ 'ye, o PCM N [M = M, II g c f 0 0.5 $. 9. FIG.5 5 Uniformly 1.oaded Plastic or Rotations at Connection. and the " rigid" connection up to a predetermined end moment. Upon reaching this pre-for the connection), a rotation capacity sufficient to de. determined end moment (My velop the yield moment at the center of the span, Af, =M, should be available. In y practical cases where the nearest available rolled section will be above the design capacity required, the excess capacity will be provided at the midspan. As in plastic redesign,end hinges only will form at the full design load. Since these hinges are designed form, =#p, more rotation capacity is required (to cross line BC). i Masonry Bearing Connections General. The AISC Specifications provide very conservative allowable stresses for bearing l on masonry and concrete which apply in the absence of Code (statutory BuildingCode) regulations (1.5.5). For all masonry laid up in mortar,most statutory codes also provide l low allowable stresses. Usually, codes distinguish among solid masonry units, bricks or block, and hollow units as well as among different classes of mortar and masonry ma-terials. Values so prescribed range in general from 50 psi to 400 psi. The AISC Specifica-annecti:n. tion is therefore seldom applicable since it includes only stone masonry,F, = 0A00 ksi, quirement is quite and brick laid in " cement" mortar,' F, = 0.250 ksi(l.5.5). 77:e authors recommend use i tanged (l.2). This of the masonry bearing values prescribed in local Codes or those recommended by na-ned to develop the tional associations dealing with masonry products. For bearing stresses on concrete, the AISC Specifications, F, = 0.35fl on the full area and F, = 0.35fl VA:/A i < 0.70f,' on the more flexible r analysis ($ = 0 at fractions of the area, utilize recent ACI Building Code (ACI 318 71) refinements for d have a very small
- The ACI Building Code is of course usually applicable under local statutory economy.
l stion at yielding of codes. For beamhearing plates and column base plates on concrete, the authors recom-mend the use of bezring values prescribed by the A CIBuilding Code. moment,M, =M,, Beam Bearing Mates. The approved design (Chapter 2, pp. 82-83, AISC Handbook) for lfg I beam bearing platesis the formula: ancept of " plastic g cad, would require I t = V3f,(n)3/F3 (Continued on page 274) l e Fig.5 5. 1 d the term " cement" can be quite property applied to a!!. il ally, the semi-rigid
- A!! mortars utilize cementitious mater a s an the & = 0 ordinate
" Supplement No. 3,1974 w . ~.
- m..
_.m c
[ ./ UQEf+ED UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD '85 zw m g g UFICE C y~w' n t,; In the Matter of ) 5E'"/ :7 ) Docket Nos. 50-445 and TEXAS UTILITIES ELECTRIC ) 50-446 COMPANY, ET AL. ) ) (Application for (Comanche Peak Steam Electric ) Operating Licenses) Station,-Units 1 and 2) ) AFFIDAVIT OF ROBERT C. IOTTI AND JOHN C. FINNERAN, JR. REGARDING THE LICENSING BOARD'S DECEMBER 18, 1984 MEMORANDUM We, Robert C. Iotti and John C. Finneran, Jr., being first duly sworn, hereby depose and state as follows: (Iotti) I am Vice President of Advanced TechnologyLfor Ebasco Services, Inc. A statement of my educational and pro-4 -'fessional qualifications was _ transmitted with Applicants ' letter-of May 16,-1984, to the Licensing Board in this proceeding. -(Finneran) I am the Pipe Support Engineer.for the Pipe Support Engineering Group at Comanche Peak Steam Electric Station. In this position, I oversee the design work of all pipe support design organizations for Comanche Peak. A statement of-my professional and educational qualifications was received into evidence as Applicants Exhibit 142B. i The purpose of-this Affidavit is to provide information regarding various issues raised by the Licensing' Board in its De' comber 18, 1984, Memorandum, e.g., sampling of torques in cinched down U-bolts at CPSES and A36/A307 steel. 6
. I. , Sampling of Cinched Down U-bolts While Applicants committed to retorque every cinched down U-bolt on single struts and snubbers to alleviate any safety questions about this issue, it was our view that the cinched down U-bolts in the field would not likely have excessive preload and would have been able to support the necessary loads. The basis-i for our position was not the sampling, but rather (1) the relaxation characteristic of the material used which would provide 'a reasonable upper bound on U-bolt preload values a'nd (2) ~ the characteristic of U-bolts at very low preload values to carry loads even if not stable in the truest sense.1 In short, the torque sampling at issue here was not reported to demonstrate that work in the field was acceptable or that no additional work needed to be done. In this regard, given the decision to retorque, the results of the testing program and finite elelsent analyses were ultimately used only to determine and provide assurance of the acceptability of the torque values to be used in retorquing. In retrospect, these results could have been obtained without taking a sample. In sum, for the ultimate use of the testing program and finite element analyses, the adequacy of the sampling is moot. i' 21 While we did not attempt to define by test or finite element analysis an absolute minimum level of preload necessary to carry load, it was our judgment that even at a very low preload value a U-bolt would be capable of supporting the necessary load. Instead of attempting to confirm this by test, Applicants opted to retorque the affected U-bolts,.as noted above.- e i ..~.... -
f 4 We maintain that reporting and using the average torque values in the sample data was appropriate because of the followings 1) the forces on the pipe and the resulting local stresses would be highest at the cross piece to pipe interface and these are determined by the combined tension in the two legs of the U-bolt, i.e., the local pipe stresses are not affected by asymmetry in the leg torques; 2) the " stability" of the configurations depends on the overall frictional resistance at the pipe cross piece and pipe U-bolt interface and this overall resistance is not affected by asymmetry in the leg torques; and 3) the difference in torques between the two legs that can exist in the field is limited and there would be a tendency for any difference to be equalized when the assembly is exposed to vibratory motion (such as due to normal vibration or earthquakes).2 We have supervised the review of construction packages associated with the sampled U-bolts and have identified 43 of the appoximately 160 U-bolts sampled which were torqued prior to October 8, 1982 and not retorqued prior to the sample.3 se. 2 The limitation in difference occurs because the difference in leg tension must be counterbalanced by the frictional resistance at the cross piece to pipe interface to satisfy l moment equalibrium. External loads of a vibratory nature momentarily reduce at each cycle the normal force and hence the frictional resistance at the cross piece to pipe interface. This reduction, in turn, reduces the difference between the unequal tension in the two U-bolt legs. 3 The construction packages of 122 of the approximately 160 U-bolts sampled were reviewed. (We limited our review to construction packages which were readily retrievable.) The date checked in each package was the date of final QC acceptance'of the associated support's initial installation. (In that QC acceptance occurs after a U-bolt is torqued, other U-bolts sampled.may have been torqued prior to October 8, 1982 and not inspected until after this date.)
- Further, the packages associated with the U-bolts torqued prior to October 8, 1982 were checked to assure that subsequent modifications had not caused the U-bolts to be retorqued.
-4' Table 1 which notes which construction packages were reviewed and which were torqued prior to October 8, 1982. (Installation of pipe supports in Unit 2 began in late 1977 and continued concurrent with pipe support installation in Unit 1 until late 1983 when virtually all of the Unit 1 pipe support installation was completed.) The torque values of such U-bolts are basically in the same range as those torqued after this date. Accordingly, there is no merit to the position that construction practices regarding torquing U-bolts were different for Unit 1 and Unit 2.4 11. Additional Information Regarding A36 and A307 Steels in response to tne Licensing Board's December 18, 1984, Memorandum (Reopening Discovery; Misleading Statement) we have reviewed previous testimony by Applicants in this proceeding to identify instances where Applicants discussed the relationship between A36 and A307 steels. We have identified three instances where Applicants discussed directly or indirectly the relation-ship between A36 and A307 steels, including the statement l referenced by the Loard in its Memorandum.5 We agree that to the l l 4 Based upon discussions with crew foremen, many of the same crews that' torqued Unit 2 U-bolts also torqued Unit 1 0-bolts. In this regard, the vast majority of the construction packages of sampled U-bolts reviewed stated the construction foremen whose crews installed the support (and, l by practice, torqued the U-bolts). Of the 45 construction foremen mentioned in the 122 construction packages reviewed, 28 still remained at CPSES. All but 3 of the 28 foremen stated that their installation crews worked in both Unit 1 i and Unit 2. 5 See (1) Affidavit accompanying Applicants' motion for (footnote continued) l l l
r 5-extent those previous statements imply that A36 and A307 steels are identical materials, they are inconsistent with our statements in our December 5, 1984, Response.6 Accordingly, we clarify those comments below and address what we believe is the reason for that inconsistency. First, however, we apologize for this inconsistency. At the time we prepared our Response we did not recall these previous statements. If we had, we would have addressed the matter in our Response. We trust that the additional information provided below will satisfactorily clarify the record. We think it is important to note, however, that correcting this apparent inconsistency does not indicate that CASE's assertions which we were addressing are correct or that any conclusions we have drawn in our motions are incorrect. As we explained in our Response, the specifications for the mechanical properties of A36 and A307 steels are different (Response at 2-3). We provided this information for the Board in our original response because the important consideration for determining representativeness is the relative mechanical (footnote continued from previous page) summary disposition regarding cinching of U-bolts (June 22, 1984), at 5 n.3, (2) Affidavit accompanying Applicants' response to CASE's answer to Applicants' motion for summary disposition regarding the effect of gaps (October 26, 1984) at 8-9, and (3) Affidavit accompanying Applicants' motion for summary disposition regarding Richmond Inserts (June 2, 1984), at 43-44.and Attachment A. 6 Affidavit of Robert C. Iotti and John'C. Finneran, Jr. accompanying " Applicants' Response to Board Memorandum (Information on Composition of A36 and A307 Steel)," December 5, 1984. j l k.
. properties. The actual chemical makeup of the steels is not, l a itself, relevant to the issues involved. However, to supplement our response, with respect to the actual chemical composition of l the steels we note that their chemical specifications are nce the l same. As can be seen in the attached ASTM specifications, there are several chemicals in A36 steel which are controlled. The chemical composition of A307 steel is, however, controlled only with respect to two chemicals, phosphorous and sulfur. Thus,. the chemical compositions of these steels are not necessarily similar. The limitations on their composition are not, however, mutually exclusive and, in fact, a steel may actually satisfy both specifications. Therefore, it is n$t accurate to state without qualification that these steels are the same. Consequently, to the extent Applicants' previous comments in this proceeding implied otherwise, they should be modified. Nevertheless, as discussed below, in certain applications the steels are, in effect, equivalent because of additional limitations imposed by specification and design requirements. In our further review of this matter we have concluded that our earlier comments on this topic were premised on a presumption, in our case held by ourselves and those working for us whom we consulted, that the materials were, in' deed, equivalent. As we demonstrated below, for a number of reasons this presumption is valid for certain application of the steels.
( ~ . In certain circumstances, by specification and/or ASME design requirements, nteels designated as A36 or A307 must have the same mechnical properties and chemical composition. In these situations the steels must satisfy the more extensive A36 chemical and mechanical property limitations, even if they are specified as an A307 steel.7 Specifically, Section 1.3 of the ASTM Specification 8 for A307 steel (Attachment A) requires that for nonheaded anchor bolts, used for structural anchorage purposes, the material shall conform to the A36 specification. (Applicants use nonheaded bolts in their Richmond Inserts but order them as A36 in the first instance.) In addition, ASME provisions governing bolting material properties require that SA307 steel used for this purpose satisfy the chemical and mechanical requirements for SA36 steel. To illustrate, we have attached applicable portions of ASME Code Cases 1644-1, 1644-4 and N-249 (which superceded 1644) (see Attachments E, F and G). The tables establishing yield strength values expressly provide that SA307 steels shall meet SA36 chemical and mechanical l 7 An exception to this situation arises with respect to A36 I headed bolts used for anchorage purposes which are also required to conform to the A307 Specification (see Section 3, ASTM Specification for A36 (Attachment A)). (See our affidavit regarding U-bolt cinching, at 5, n. 3 (cited by the Board in its Memorandum, at 4).) 8 The ASTM specification is applicable to bolts used in Richmond Inserts. The ASME specifications for A36 and A307 (under which these steels are designated SA36 and SA307) are essentially equivalent to the ASTM specifications (Attachments C and D). The "A" and "SA" prefixes have also been used interchangeably in this proceeding. L
(. . requirements.9 Thus, for design purposes, SA36 and SA307 bolts which are governed by these requirements may be considered " equivalent". In summary, the presumption that A36 and A307 steels are " equivalent" is based on restrictions imposed by both specifications and design requirements. Again, however, the steels are not the same but only equivalent in certain contexts.. We wish to reemphasize that Applicants use only A36 cteel both in U-bolts and bolts used in Richmond inserts.lO CASE's repeated assertions to the contrary are incorrect. We hope'the additional information we have provided regarding both A36 and A307 will assist in clarifying the subject for the Board. The Board commented in its Memorandum (at 5) that: Applicants' tests related to friction, stiff-ness, relaxation and creep, characteristics of steel that are not readily assertained from data on yield and tensile strength. We are not certain what the board's precise concern is from this comment. We surmise that the Board also is interested in information to demonstrate the representativeness of steels in the field with respect to these properties. Accordingly, we -provide information below to respond to that concern. 9 We recognize that there'are differences in the manner in ~ which requiremente for different grades of A307. steel are established under the Code Cases. We do not address those differences here because they are not relevant tx) Applicants' practices at issue. 10 A single exception to the use of A36 steel in. bolts used in Richmond Inserts was noted in our affidavit regarding Richmond Inserts (at 9). W
[ ,c.
- ,r r-UNITED STATES OF AMERICA M 18 A10:3 NUCLEAR REGULATORY COMMISSION LFFIC[cUCCdij%i$$.
BEFORE THE ATOMIC SAFETY AND LICENSING BOARD M A scy In.the Matter of }{ I }{ TEXAS UTILITIES ELECTRIC }{ Docket Nos. 50-445-.1 i COMPANY, et al. }{ and 50-446-y V (Conanche Peak Steam Electric }{ ' Station, Units 1 and 2) }{ CERTIFICATE OF SERVICE By my signature below, I hereby cerrify that : rue and correct copies of CASE'S 4th Motion for Summary Disposition: to Disqualify the Use oflSA307 and SA36: Threaded Parts have been sent to the names listed below this 14th day of January ,198_5, by: Express Mail where indicated by
- and First Class Mail elsewhere.
i
- Administrative Judge Peter B. Bloch
- Nicholas S. Reynolds, Esq.
U. S. Nuclear Regulatory Commission Bishop, Liberman, Cook, Purcell 4350 East / West Highway, 4th Floor & Reynolds Bethesda, Maryland 20814 1200 - 17th St., N. W. Washington, D.C. 20036
- Judge Elizabeth B. Johnson Oak Ridge National Laboratory
- Geary S. Mizuno, Esq.
P. O. Box X, Building 3500 Office of Executive Legal Oak Ridge, Tennessee 37830 Director U. S. Nuclear Regulatory
- -Dr. Kenneth A. McCollom, Dean Commission Division of Engineering, Maryland National Bank Bldg.
Architecture and Technology - Room 10105 Oklahoma State University 7735 Old Georgetown Road t Stillwater, Oklahoma 74074 Bethesda,-Maryland 20814-i
- Dr. Walter 9. Jordan Chairman, Atomic Safety and Licensing Carib' Terrace Motel 3oard Panel 552 N Ocean Blvd.
U. S. Nuclear Regulatory Commission
- Pompano Beach, Florida 33062 washington, D. C.
20555 1 bi
m 1 Chairman Renea Hicks, Esq. Atomic Safety and Licensing Appeal Assistant Attorney General Board Panel Environmental Protection Division U. S. Nuclear Regulatory Commission Supreme Court Building
- fashington, D. C.
20555 Austin, Texas 78711 John Collins Regional Administrator, Region IV U. S. Nuclear Regulatory Commission 611'Ryan Plaza Dr., Suite 1000 Arlington, Texas 76011 Lanny A. Sinkin, Director of NIRS 1346 Connecticut Avenue N.W., 4th Floor Washington,.D. C. 20036 Dr. David H. Bolt: 2012 S. Polk Dallas, Texas 75224 Michael D. Spence, President Texas Utilities Generating Company Skyway Tower 400 North Olive St., L.B.-81' Dallas, Texas 75201 Docketing and Service Section. j -(3 copies) office of the Secretary U. S. Nuclear Regulatory Commission Washington, D. C. 20555 A E> _p.) Juanita Ellis, President CASE (Citizens Association for Sound Energy) 1426 S. Polk Dallas, Texas 75224 214/946-9446 2}}