ML20088A702
| ML20088A702 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 04/11/1984 |
| From: | Horin W, Reynolds N BISHOP, COOK, PURCELL & REYNOLDS, TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC) |
| To: | |
| Shared Package | |
| ML20088A703 | List: |
| References | |
| NUDOCS 8404120423 | |
| Download: ML20088A702 (22) | |
Text
N bk s
a DOCKETED April 11, 1984 l
UNITED STATES OF AMERICA 84 APR 12 A9:37 NUCLEAR REGULATORY CCMMISSION r tr-
. r.,,
BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of
)
)
Docket Nos. 50-445 and TEXAS UTILITIES ELECTRIC
)
50-446 COMPANY, -- --e t al.
)
)
(Application for (Comanche Peak Steam Electric
)
Operating Licenses)
Station, Units 1 and 2)
)
APPLICANTS' RESPONSE TO PARTIAL INITIAL DECISION REGARDING A500 STEEL In a Partial Initial Decision regarding revisions to the material properties tabulated for A500 tubular steel the Licensing Board ordered Applicants to submit analyses demonstrating that pipe supports manufactured with A500 steel have adequate safety margins.1 Pursuant to that order, Texas U tilities Generating Company, e t al.
(" Applicants") hereby submit their response to the Board's Decision.2 For the reasons set 1
Partial Initial Decision (Change in Material Properties for A500 Steel) (October 6, 1983)
(" Decision"), at 8.
2 The Board has since issued its Memorandum and Order (Quality Assurance for Design), December 28, 1983, in which it addressed several pipe support design allegations and requested that Applicants submit a plan to resolve certain concerns raised in the Memorandum and Order.
The Board subsequently questioned whether it would be profitable to receive evidence regarding the A500 steel question independent of "some additional procedure such as the independent design review [the Board] suggested."
(Memorandum and Order (Additional Scheduling Order), January 3, 1984 a t 6).
However, in the January 12, 1984 conference
( footnote continued) 8404120423 840411 i
eDR ADOCK 05000445 i
gso a
roa
t h forth below, we urge the Board to find that Applicants correctly assessed the impact of the revision to A500 tube steel yield values and that pipe support designs which utilize A500 tubular steel provide adequate margins of safety and raise no concern regarding the safety of the facility.
I.
BACKGROUND Allegations regarding Applicants' pipe support designs were litigated over two hearing sessions (July and September, 1982),
and were addressed in hundreds of pages of briefs and testimony by all parties, prior to the May, 1983 hearing session, at which the NRC Staf f's evaluation of those allegations was presented.
During those hearings the intervenor raised for the first time a question regarding the effect a revision to the yield strengths published (in 1981) by the ASME for A500 tube steel could have on pipe support designs for Comanche Peak.3 The intervenor, argued that the reduction in yield strengths reduced the margin of safety for pipe support designs using A500 tube steel.
(Tr.
i 6789-92.)
The Board immediately heard arguments from the parties regarling the applicability to Comanche Peak of the ASME Code Case vbich revised the subject yield strengths.
Following those (footnote continued from previous page) call the Board encouraged Applicants to file this response without awaiting presentation of further evidence on pipe support design allegations (Tr. 9218).
4 3
By Code Case N-71-10, the ASME reduced the yield values for A500 tube steel.
This Code Case was-approved by the NRC for use in the design of components for nuclear facilities by Regulatory Guide 1.85, Revision 19 (April 1982) (see discussion infra,Section III.B.).
I 4 arguments, the Board ruled that the NRC had exempted from J
l application of the Code Case those components which had been contracted prior to the effective date of the code Case.
Consequently, the Board cut off further presentation of evidence regarding the effect of those revisions on pipe support designs.
(Tr. 6803, 6806-09, 6816.)
However, the Board later determined that it is necessary that Applicants analyze the effect of the revision to A500 steel yield strengths on the pipe support designs at Comanche Peak in order to assure that adequate margins of safety exist in those designs (Decision at 8).
i II.
SUMMARY
OF ARGUMENT The Board determined that NRC regulations viz., General Design Criteria ("GDC") 1 and 4, and 10 C.F.R.
650.55a, require that Applicants perform specific detailed analyses to address the ef fect of the revised A500 tube steel yield strengths.
(Decision at 7-8.) Applicants believe, however, that neither NRC regula tions nor guidance require that detailed analyses be performed to account for the revised yield strengths.
Applicants believe that upon reassessing applicable NRC regulations and guidance materials, the Board will agree that those provisions may be fully satisfied with respect to the changes in yield strength values without conducting the particular detailed analyses the Board asks be performed.
Applicants demonstrate below that they, in fact, considered (even before the NRC authorized use of the revised yield values) the impact of the
e c 4 revision on their pipe support designs and, for several reasons, concluded that adequate margins of safety remained in those designs.
Applicants submit that Applicants' consideration of these factors satisfied applicable provisions of NRC regula tions and guidance materials.
In addition, to confirm the soundness of l
their original conclusions and to assist the Board in resolving this matter, Applicants have performed analyses of the effect the J
l use of the revised yield strengths would have on pipe support I
designs at Comanche Peak.
These analyses confirm that i
Applicants' initial determination was correct and that no safety concern is raised by the use of the original yield strengths in l
Applicants' pipe support designs.
III. REANALYSIS OF PIPE SUPPORT DESIGNS IS NOT REQUIRED BY COMMISSION REGULATION OR GUIDANCE TO ASSESS THE IMPACT OF THE REVISION OF A500 TUBE STEEL YIELD VALUES A.
Commission Regulations Do Not Require Reanalysis of Pipe Support Designs to Address Revised Yield Values The Board found that Applicants have " failed to demonstrate compliance with General Design Criteria (GDC) 1 and 4 in the design of pipe supports using A500 Steel at Comanche Peak."
i Specifically, the Board _ concluded that Applicants have "not demonstrated that welded supports using A500 Steel have been i
designed with adequate safety margins."4 Although recognizing that Applicants sa'tisfy ASME Code provisions governing this matter, the Board interprets GDC 1 and 4 to require Applicants to i
4 Decision at 2-3 (footnote omitted).
e o
t )
a perform analyses of the effect of the revised material properties, concluding that Applicants can satisfy those criteria "only" by performing analyses which seek " extreme or limiting cases" and which quantify "the combined effects of errors in code values and other variations typically covered by safety factors" to show "that there are adequate safety margins remaining."5 For the reasons set forth below, Applicants believe, and are confident the Board will agree, that neither the General Design Criteria or other Commission regulations require that such reanalyses be performed.6 1.
General Design Criteria 1 and 4 Neither GDC 1 nor GDC 4, by their own terms, mandate that a particular reanalysis of pipe support designs be performed to assess the subject revisions to material properties.
The broad provisions of GDC 1 and 4 are, as their name implies, general requirements for the design of nuclear power reactors.
These general requirements are " intended to provide engineering goals 5
Decision at 4-5, 7-8.
6 The Board also determined that the contract date for the piping on which the support is placed governs Code Case applicability.. Decision at 3.
For the Board's information, we note that the ASME Code also provides an additional criterion for determining code Case applicability, viz.,
mutual consent of owner, manufacturer and installer.
(See NA-ll40(f); see also ASME Interpretive Letter to Texas Utilities (November 18, 1983), Finneran Affidavit, Attachment).
. )
l rather than precise tests or methodologies by which reactor safety [can] be fully and satisfactorily gauged."7 As the Commission has stated, they are cast in broad, general terms and constitute the minimum requirements for the principal design criteria of water-cooled nuclear power plants.
There are a variety of methods for demonstrating compliance with GDC.
Through regula tory guides, standard format and content guides for safety analysis reports, Standard Review Plan provisions, and Branch Technical Positions, license applicants are given guidance as to acceptable methods for implementing the general criteria.
However, applicants are free to select other methods to achieve the same goal.
[ Petition for Emergency and Remedial Action, CLI-78-6, 7 NRC 400, 406 (1978).]
Consistent with the Commission's assessment, GDC 1 and 4 are models of generality.
GDC 1 reads, in applicable part, as follows:
Structures, systems and components important to safety shall be designed, fabricated, erected, and tes ted to quality standards commensurate with the importance of the safety functions to be performed.
Where generally recognized codes and standards are used, they shall be identified and evaluated to determine their applicability, adequacy, and sufficiency and shall be supplemented or modified as necessary to assure a quality product in keeping with the required safety function.
Similarly, GDC 4 reads, in applicable part, as follows:
Structures, systems and components important to safety shall be designed to accommodate the effects of and to be compatible with the environmental conditions associated with normal operation, maintenance, testing and postulated accidents, including loss-of-coolant accidents.
7 Nader v. NRC, 513 F.2d 1045, 1052 (D.C. Cir. 1975).
4 4 In short, neither criterion establishes particular standards or requires that specific analyses be employed in the design and construction of a nuclear reactor.
Even the above-cited portion o f GDC 1, which provides for evaluation of generally recognized codes and standards to assure their " applicability, adequacy, and sufficiency," does not dictate that a particular method of evaluation be undertaken to assure a quality product.
Accordingly, Applicants submit that it is not correct to conclude that Applicants can "only" satisfy GDC,1 and 4 "by analyzing the e f fect of this [ revision to yield strengths] on its plant and showing there are adequate safety margins remaining,"8 seeking
" extreme or limiting cases" and quantifying "the combined effect or errors in code values and other variations typically covered by safety factors".9 In short, the general provisions of GDC 1 and 4 do not support such a confined approach.
2.
10 C.F.R. 650.55a The Board also relies on a portion of 10 C.F.R.
50.55a to support its interpretation of the GDC requirements.
The Board states tha t 10 C. F.R. $50.55a(d)(2)l0 " lends further substance" 8
Decision at 4-5, 7-8.
Applicants agree with the Board's additional conclusions on this matter, viz., that the 3
Commission did not intend that Applicants rely entirely on Code provisions or " erroneous" Code sections to assure safety.
As will be shown later, Applicants have not and do not place such reliance to assure a safe facility.
9 Decision at 7.
10 This citation appears to be a typographical error, in that the portion of 10 C.F.R.
50.55a(d) regarding the (footnote continued)
. to the GDC requirements by specifying the applicability of the ASME Code.
However, the " substance" added to the GDC by this requirement is not specified.
Nor does any portion of 10 C.F.R.
$50.55a require that a detailed reanalysis be done.
Accordingly, we submit that neither 10 C.F.R.
50.55a (or GDC 1 and 4) supports a finding that specific reanalysis of pipe support designs.is required to assess the impact of the revised yield values.
B.
Commission Guidance Does Not Require That Reanalyses of Pipe Support Designs be Performed to Address Revised Yield Values As the Board has found, the principal Commission regulation governing the consideration of generally recognized codes and standards in the design and construction of nuclear power reactors is GDC 1 of 10 C.F.R. Part 50, Appendix A.
As noted above, the general design criteria establish only general goals and do not specify the particular tests or methodologies to be employed to achieve those goals.ll Indeed, the Commission recognizes that general design criteria are cast in broad, general terms, and that additional guidance is necessary to establish means, acceptable to the NRC, for implementing those (footnote continued from previous page) construction of piping which is applicable to Comanche Peak is 10 C.F.R.
50.55a(d)(3), by virtue of the December 19, 1974 date of issuance of the construction permits for Comanche Peak Units 1 and 2.
11 See Nader v.
NRC, supra, 513 F.2d at 1045.
. criteria.12 Regarding the situation at hand, the Commission has established guidance as to the proper means for implementing the provisions of GDC 1 with respect to the consideration of ASME Code Cases.
Regula tory Guide 1.85, " Materials Code Case Acceptability, ASME Section III, Division 1," establishes means acceptable to the Commission for utilizing ASME Code Cases (oriented to materials and testing) in the licensing of light-water-cooled nuclear power reactors.
Regulatory Guide 1.85 specifically references the portion of GDC 1 which provides for the use of generally recognized codes and standards.
Thus, by its own terms, Regulatory Guide 1.85 establishes a generally accepted methodology for satisfying GDC 1 with respect to the means by which materials and testing data contained in ASME Code Cases are to be utilized.13 Regulatory Guide 1.85 does not suggest that detailed analyses be performed to address revised material properties published in ASME materials Code Cases.
In fact, as the Board has already acknowledged, this guide expressly exempts components ordered to a particular Code Case that is subsequently revised or annulled from the provisions of the revised Code Case.14
- Thus, 12 See Petition for Emergency and Remedial Action, supra, CLI-78-6, 7 NRC a t 406.
t 13 Regula tory Guide 1.85, Section A, Introduction.
14 Regulatory Guide 1.85, Section D, Implementation; Tr. 6803, 6806-09, 6816.
f r-..
. this guide does not support the conclusion that specific analyses need be performed with respect to the effect of Code Case revisions.15 C.
The Board Should Reconsider Its Interpretation of the Record Regarding the Potential Effect of Revised Yield Values on Factors of Safety In assessing the impact that the revision to the A500 tube steel yield values could have on the pipe support designs at Comanche Peak, the Board concluded that utilizing the original yield values may give rise to a design deficiency with respect to consideration of cyclic stresses.
In reaching this conclusion, the Board interpreted Applicants' testimony regarding factors of
)
safety and the criteria which govern consideration of cyclic stresses in the design of linear pipe supports.
However, as 4
demonstrated below, that testimony does not suggest that the reduction of published yield values for A500 steel should.give rise to a concern regarding the adequacy of Applicants' pipe support designs with respect to cyclic stresses.
15 It is noteworthy that Regulatory Guide 1.85 states that "all published Code Cases in the area of materials and testing that are applicable to Section III of the Code and were in effect on [a date certain for each revision to the Guide],
I were reviewed for inclusion in this guide."
(Regulatory
~
Guide 1.85, Section B (emphasis added).)
Indeed, although specific conditions have repeatedly been attached to the use of the very Code Case at issue, no limitations have been imposed on the use of the original A500 tube steel yield strengths, even after their revision (Regulatory Guide 1.85, Section C (see e.g. Revisions 10-21))..
. The Board concluded that one of Applicants' witnesses, Mr.
Reedy, relied on " safety factors based entirely on ultimate" in explaining why the revision to the A500 yield strengths raised no safety concern.
(Decision at 6).
Actually, Mr. Reedy testified that the ASME Coda employs a dual safety factor, with criteria established both on yield and ultimate strengths (Tr. 6923-24).16 In addition, a crucial aspect of Mr. Reedy's testimony which the Board should note is that under this dual system it is, in fact, more conservative in most instances to employ factors of safety based on ultimate strengths (Tr. 6925-26), and that it is not correct to associate (as did the intervenor's witnesses) stresses greater than yield with failure or collapse of the component (Tr.
6924-25).17 In fact, the record reflects that a margin of safety on yield is used in the ASME Code (Tr. 6923) and in particular in the design of linear pipe supports (see ASME Code, Appendix XVII-2000, Sections 2211-2216).
In short, the record does not suggest, nor have Applicants' witnesses advocated, that the ASME j
Code relies on " safety factors based entirely on ultimate."
16 The Board recognized in one portion of its Decision that Mr.
Reedy testified that the ASME Code employed dual safety factors (Decision at 6).
17 It was in this context that Mr. Reedy alluded to Mr.
Vivirito's testimony, regarding application of elastic analyses to material not strictly within the elastic range (Tr. 6924-25 (regarding cyclic stresses and " shakedown" to elastic action)), to illustrate his own testimony that it was incorrect to conclude that elastic analysis was valid only within the yield stress range and that once beyond that range collapse of the component would occur (Tr. 6925).
In this regard, the Board should also reconcider its " understanding that if yield is exceeded, plastic analysis (that applicant has not done) would be required" (Decision at 7 (emphasis added); compare Tr. 6927-30.)
e Further, the Board should reassess its interpretation of the evidence of record regarding consideration of cyclic stresses.
Specifically, the Board concludes that the limitations on cyclic stresses are premised on yield strength values themselves (Decision at 6).
However, as Applicants' witness Mr. Vivirito testified, the Code imposes a stress limit of twice yield whenever it requires consideration of repetitively applied, i.e.
cyclic, secondary stresses in order to assure that " shakedown" to elastic action occurs (Tr. 5893 See also Tr. 7085-86, Applicants' Exhibit 142 a t 18-19).
Thus, it is not correct to conclude that stress limits based solely on yield strength would approach the limit established for cyclic stresses, viz., twice yield.
In addition, in order to dispel the Board's concern in this area, the Board should note that Applicants have not taken advantage of the Code provision which permits allowable stresses for cyclic loads to be set at twice yield (ASME Code Section NF-3231.l(a)).
Instead Applicants' pipe support design guidelines establish allowable stress levels for these conditions at values less than yield.
Thus, the small revision to the yield values for A500 steel does not give rise to a concern with respect to the consideration of cyclic stresses.
(Finneran Affidavit at 2.)
Finally, the Board should review its conclusion that
" assuming that some errors are covered by the safety factors, there is no testimony about the magnitude of errors intended to be covered" (Decision at 7).
Mr. Reedy's testimony that even a decrease of 20 percent in yield strength of A500 tube steel would
i e
e not have a significant effect on safety or require items to be taken out of s'ervice (Tr. 6809-10) is directly on point.
There is no testimony contradicting Mr. Reedy's assessment.
- Further, as will be discussed in more detail below, the ASME recently issued an interpretive letter to Texas Utilities reaffirming i
their judgment (and thereby confirming Mr. Reedy's testimony)
~
that the original A500 tube steel yield values romain adequate.
(See Attachment to Finneran Affidavit, Reply 2.)18 In view of the foregoing, the Board should revise its decision, and find instead that Applicants have demonstrated (see also discussion infra) compliance with GDC 1 and 4 and that there is no basis for concern with respect to the impact of the revised A500 tube steel yield values on pipe supports at Comanche Peak.
IV.
APPLICANTS' USE OF A500 YIELD STRENGTH VALUES SATISFIES NRC REGULATIONS AND GUIDANCE AND RAISES NO SAFETY CONCERN A.
Applicants' Properly Concluded That No Safety Concern Existed Ifith The Use of Original Yield Strength values As discussed above, NRC regulations and guidance do not require that a particular analysis or evaluation be performed to account for tha revised A500 yield strength values.
Nonetheless, even prior to the approval of Code Case N-71-10 by the NRC, Applicants recognized that the yield strength values of A500 18 In accordance with the procedures applied by the ASME to the development of interpretive letters, Mr. Roedy was not allowed to participate in the deliberation or to vote on this i
interpretive letter because of his involvement in this proceeding.
I
e tubular steel had been slightly decreased.
Applicants were aware, however, of several factors in their pipe support designs which assured that no adverse safety impact would result from their use of the original yield strengths for A500 tube steel.
Applicants properly concluded that it was unnecessary to reanalyze the pipe support designs at Comanche Peak to account for the use of the original yield values of A500 tubular steel.
Each of the factors known to and considerea by Applicants in reaching that conclusion is discussed below.
1.
The ASME practice When Applicants first considered the significance of the revised yield values set forth in Code Case N-71-10, Applicants recognized that the ASME thoroughly reviews all Code Cases before their issuance to assure that no potential safety concerns are raised by prior practices which may be affected by new og revised Code Cases.
In situations where such concerns may exist, the ASME will either make a Code Case mandatory or notify all parties who may be affected, of the potential safety concern.
In this instance, the ASME did not make the subject Code Case mandatory and did not issue a noti.ce of a potential safety concern.
Accordingly, Applicants were confident that the use of the original yield values for A500 tubular shape steel presented no safety concern.
( Af fidavit of John C.
Finneran at 3. )19 19 To confirm the soundness of their initial assessment of the ASME's actions, Applicants requested and received from the ASME an interpretive letter regarding these revised values.
( footnote continued)
e l i
2.
Applicants' conservative design process Applicants did not, however, base their conclusion regarding i
the use of yield values for ASOO steel solely on the ASME determination described above.
Applicants also recognized that several conservatisms provide additional margins of-safety that
~
more than offset the small reduction in A500 tube steel yield values occasioned by Code Case N-71-lO.
First, Applicants knew that design criteria other than the tube-steel allowable stress values are the limiting design considerations (i.e.,
criteria i
whose design limits are most' closely approached under calculated loading conditions) for pipe supports at Comanche Peak.
Thus, by l
i assuring that those criteria are met in the first instance, i
Applicants have a high level of assurance that the stresses in the tube steel will be well below the allowable stresses.
For example, adherence to the 1/16 inch deflection criteria alone, which has been extensively addressed in this proceeding,,will in 1
most instances limit the stresses imposed on the tube steel to levels well below the allowable stress values.
In addition, the stress limits on Hilti (and other anchor) bolts also serve as a controlling design consideration.
Given such conservatisms in their design process, Applicants were satisfied that there was a i
high level of assurance that stress levels in tube steel members i
j i
(footnote continued from previous page)
In that letter, the ASME states that because of the many I
safety factors and design constraints applied to the yield strength values used in the design of piping supports, the ASME determined that no potential safety concern was raised by the revision and, thus, it was not necessary to provide special notice regarding. the change.
(Finneran Affidavit, j
Attachment.)
?
. i would not exceed the revised A500 tube steel allowables under calculated loading conditions.
Applicants' conclusion in this regard is confirmed by the analysis of actual supports at Comanche Peak, discussed below.
(Finneran Affidavit at 3-4.)
In addition to the above, further assurance that stresses in tube steel members will not approach the revised allowables for A500 tube steel is derived from the fact that Applicants' pipe support design organizations use very conservative design i
philosophies.
For example, Applicants' support designers frequently apply more conservative allowable load levels than are required.
In this regard, Applicants. apply level B stress allowables in assessing the more severe level C loads.
- Further, each pipe support design organization, as a general practice, employs stronger tube steel sections than necessary in ceder to
, provide a contingency for possible changes in support loads and stresses.
This fact is evidenced by the analysis of actual supports discussed below.
Thus, the stresses imposed on tube steel members in Applicants' pipe support designs are most often much less than the stress limits imposed by A5ME criteria'and NRC regulations.
(Finneran Affidavit at 4-5. )
3.
Actual material properties j
In addition to the above, Applicants' experience with test data (which are received with each shipment of steel) has been that the actual yield strengths are generally substantially greater than the published values.
These test data are provided 1
6 3,-
y-,
v
- .,em
/
e-
=
_ 17 _
to Applicants by fabricators of steel materials who perform ma terial tests for various properties, including yield strengths, t
on each productio'n run of material.
This fact also contributed to Appl _icants' confidence that the reductions in A500 tube steel
~
yield strengths r'aised no concern for the ultimate safety of pipe support designs a't Comanche Peak.
To confirm Applicants' judgment on this matter, a worst case analysis was performed of actual supports at Comanche Peak.
(Finneran Affidavit at 5-6.)
j This analysis is discussed'below.
In sum, Applicants' design process has several design conservatisms that are not formally accounted for or relied upon l
In routine design practice.
Applicants were aware of such conservhtisms at the time the A500 yield strengths were revised and, thus, wero; satisfied.that the Adequacy of designs using the original yield v'alues was assured.
It is wholly appropriate that Applicants considered such factors in this instance to assure that the use of the original yield values did not give rise to a concern for the adequacy of the overall support designs.
Applican'ts submit! that the above-described assessment fully
~
satiafied the requirements' of GDC 1 and 4 as well as the guidance established in Regulato'ry Guide 1.85.
Accordingly, the Board
~
should find that. Applicants'have satisfied applicable NRC regulations, and in;particular GDC 1 and 4, and NRC guidance set
]
forth in Regula tory Guide l'.85, with respect to consideration of revised yield strength values for A500 tube steel set forth in ASME Code Case N-71-10.
> B.
Applicants' Evaluation of Effect of Revised A500 Tube Steel Yield Values on Pipe Support Designs Confirms That Adequate Margins Of Safety Remain With original Yield Values Applicants firmly believe that the information presented above adequately demonstrates that the revision to the A500 tube steel yield values raised no concern for the adequacy of the pipe support designs and that applicable NRC regulations and guidance are satisfied by the use of the original values.
Nevertheless, to respond to the Board's concerns, Applicants performed an analysis of the actual impact of the revised values on support designs.
1.
Sampling of supports demonstrates insignificant impact of revised yield values i
To assess the impact of the revised A500 tube steel yield values on actual pipe support designs at Comanche Peak, and to confirm Applicants' original conclusion that no safety concern l
exists with respect to those revised values, Applicants developed a sample of a large number of supports which were evaluated using the revised yield values.
This analysis demonstrates that all stresses in the sampled tube steel support members remained below even the reduced allowable stresses and that, in fact, the stresses in the vast majority of supports remained significantly below those allowables.
(Finneran Affidavit at 6.)
To generate the sample of supports for this analysis, 1
Applicants first developed alphabetical lists of all Unit 1 and common area ASME supports for each support design organization
l (NPSI, ITT-Grinnell and PSE).
ITT and NPSI then selected from the lists every hundredth support, and PSE selected every ninetieth.
If the selected support did not utilize A500 tube steel, the next support on the list which did was chosen.
The resulting selection of 182 supports contained an excellent mix of buildings, systems, classes and types of supports.
Each selected support was then reviewed to determine the effect of substituting the reduced allowable stresses for the A500 tube steel based on the yield strengths published in Code Case N-71-10.
Applicants then tabulated the percentage of the allowable stress seen by the most stressed tube steel member of each support.
The results of this tabulation are, as follows:
Percentage of reduced allowable stress for highest Number of Supports stressed tube steel member 82 0-10 52 10.01-20 22 20.01-30 9
30.01-40 8
40.01-50 3
50.01-60 4
60.01-70 1
70.01-80 0
80.01-90 l'
3[f IIT Analysis of the above sample demonstrates that for over 95% of the sampled supports the tube steel member seeing the maximum stress is stressed less than 50% of the reduced A500 allowable.
In fact, the average percentage of tha allowable stress seen by those members is 16.4%.
None is stressed to the reduced allowable.
(Finneran Af fidavit at 7.)
t
o o
20 -
In sum, there is a high level of confidence that supports at Comanche Peak utilizing A500 tube steel will not be stressed at levels which exceed the allowable stress derived from the reduced yield values.
This analysis also confirms Applicants' initial judgment that the design process for pipe supports is very conservative, including with respect to tube steel support members.
Accordingly, there is a high level of assurance that an adequate margin of safety exists with respect to the stresses imposed on tube steel members.
2.
Actual material properties of A500 tube steel significantly exceed published yield values As already noted, the actual strength of fabricated steel i
ma terials is generally much greater than the published material properties (e.g. yield strengths) used for design calculations.
To demonstrate the implications of this fact for the pipe support designs at Comanche Peak which use A500 tube stoel, Applicants selected for further analyses the ten highest stressed support members from the sample of supports discussed above.
Applicants determined from the certified mill test reports for these support members that the minimum actual yield strength for the A500 tube steel material in these members was 56.3 ksi.
To account for the 15% reduction in yield strengths assumed by the ASME to result
-from welding, we reduced the lowest of these actual yield strengths by 15% to 47.8 ksi.
Yet, this value is still 33% above the 36 kai yield strength tabulated by the ASME ~ for A500 tube steel in Code Case N-71-10, and still greater than the 42 kai l
l
e 21 -
originally used.
Thus, even this comparison of actual v.
published yield strengths for A500 tube steel used in more highly stressed supports at Comanche Peak demonstrates that there is a significant favorable margin between these values.
This fact further emphasizes the highly conservative nature of the support designs.
This evaluation also confirms Applicants' initial judgment at the time the A500 tube steel yield values were reduced that consideration of actual (rather than published) material properties would more than offset the slight reduction of the yield values published by the ASME.
(Finneran Affidavit at 7-8.)
C.
Revised Yield Values Are Conservative A final point to recognize is that the revised yield strengths for A500 tube steel are themselves conservative values selected by the ASME.
The ASME recognized that the yield strength of A500 steel in the cold wrought condition may be slightly reduced in the heat affected zone of weldments.
Accordingly, the values the ASME selected are conservative values for the yield strengths for A501 and A36 materials.
(Finneran Affidavit at 8-9.)
Accordingly, the Board should conclude that
t D the values selected by the ASME are conservative in themselves and, thus, provide further assurance that no significant sa fe ty concern is raised by use of the original values.
Respectfully submitted, t
lad w SllMil Nichgla s S'. Reth61ds
) )
Wi Ciam A.
Hor'in' Counsel for. Applicants BISHOP, LIBERMAN, COOK, PURCELL & REYNOLDS 1200 Seventeenth Street, N.W.
Washington, D.C.
20036 (202) 857-9817 April 11, 1984 t
..