ML20197H322
| ML20197H322 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 06/17/1984 |
| From: | Finneran J TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC) |
| To: | |
| Shared Package | |
| ML20197H236 | List: |
| References | |
| NUDOCS 8406180392 | |
| Download: ML20197H322 (35) | |
Text
\
June 17, 1984 UNITED STATE 9 OF AMERICA NUCLEAF REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of )
) 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 JOHN C. FINNERAN, JR.
REGARDING STABILITY OF PIPE SUPPORTS AND PIPING SYSTEMS I, John C. Finneran, Jr., hereby depose and state as follows: I am employed by Texas Utilities Generating Company as Project Pipe Support Engineer for the Comanche Peak Steam Electric Station. In this position I oversee the design work of all piping support design organizations at Comanche Peak. A statement of my educational and professional qualifications is in evidence as Applicants' Exhibit 1428.
Q. What is the purpose of your affidavit?
A. This affidavit addresses allegations raised by CASE regarding the stability of individual supports and questions posed by the Board regarding the process by which Applicants identified and corrected potentially unstable supports at Comanche Peak. CASE's allegations and the Board's questions are set out in the Board's Memorandum and Order (Quality 8406180392 840617 PDR ADOCK 05000445 0 PDR
O O
Assurance For Design), December 28, 1983, at 22-23, and its Memorandum and Order (Reconsideration Concerning Quality Assurance For Design), February 8, 1984, at 15-17, 19-22.
As demonstrated below, Applicants had identified in May, 1981, as potentially unstable, types of supports later argued in the hearings by Messrs. Walsh and Doyle to be unstable. Applicants modified those categories of supports determined to warrant improvement of their stability in the normal course of their design process. I also discuss below the reasons why some of the supports that CASE alleged to be unstable are, in fact, stable and require no modifications.
Q. There has been much discussion in these hearings about
" stability." Would you please describe what stability is in the context of piping and support design?
A. ASME Code Section NF, Appendix XVII,Section XVII - 2221(a),
provides as follows:
general stability shall be provided for the structure as a whole, and for each compression element.
When the Code talks about stability, it simply means that the support should be capable of supporting its applied load. If a system of supports is incapable of supporting the piping system under its various loading conditions, the system is unstable. Dr. Bjorkman concisely described in the April 1984, hearings what instability meant. He defined instability, as follows:
l i
3-Loss of complete, total load-carrying capability. That's instability. Instability can result from many - or be arrived at in many ways - from column buckling, rigid body mode bending, the bending of the cantilever beam when you exceed the yield hinge, the plastic hinge. That's instability in collapse.
So instability takes many forms, but it results in the same thing -- collapse."
[Tr. 13022]
Thus, there is not just one form of instability. If a column fixed at one end were loaded wi'th enough. axial. load to cause collapse of the column, it.would be unstable. For -
the instability tx) occur, some deformation and bending of the column are required. The type of stability.most discussed in these hearings is called rigid body' instability. This is the stability _ question related to the rotation-of the pipe' attachment (frame, clamp,' snug U-bolt) around the pipe. No real overload condition or member
" bending" need occur for the support to be unable .tx) carry its intended load. Thus, the name rigid. body instability.
To illustrate-this condition, we shall start with a
~
single structural support as shown in Figure 1(a), ,
consisting of a single strut, loaded in compression by load F, and pinned at the building support. It is not a pipe-support; it-is simply a structure intended to' resist load F.
]
i v l l
I l
w .4-.. e p-*4 4 m T8 :w
4-l '
f /J ] L I y
////// I 1 I I g
M '
u
,\ L i s\ !
\ s l.
\ \
\% l li h f
L F
(A) (B) (C)
Figure 1 with a small eccentricity of the applied load, the strut could swing to the side and become incapable of supporting the applied load F. Therefore, it is an unstable support. This would be rigid body instability.
In Figure 1(b), the load F is replaced with a pipe and pipe clamp and the clamp is snugged tight on the pipe. In theory this support is still potentially unstable because a small eccentricity could cause the strut to swing to the side and become incapable of supporting the applied load, as shown in Figure 1(c).
In Figure 2, the type of structure illustrated in Figure 1(b) is placed in a piping system where the pipe is supported in all directions by other supports. 1 l
1 i
l
1 4 ,
jj1 1 'L 7 3 support #2 support i1 A I
4
- 1 i
1 Figure 2 In this figure it can readily be seen that if a compressive load were applied to support #1, the support would behave in a stable manner because the pipe cannot swing to the side (compare Figure 1(c)) as it is being restrained in that direction by support #2. Support #1 provides the same stabilizing effect for Support #2.
It can therefore be seen that a support by itself (Fig.
1(b) and 1(c)) reay appear potentially unstable, but when it is connected to the piping system (Fig. 2), it is in fact stable, and the piping system is stable. Dr. Chang, Applicants' Chief Engineer, Pipe Support Engineering,
l I
testified regarding the stability of supports and the need to consider system stability in September, 1982 (Applicants' Exhibit 142 at 28):
Q. Is it necessary that each pipe support structure be stable by itself?
A. (Chang) No, it is not. The important thing is that the piping and supports as a system are stable.
Similarly, Dr. Chen, one of the NRC Staff witnesses regarding the pipe support design issues, addressed -
stability in the SIT Report, at p. 27 as follows:
The question of whether a particular support is stable or unstable when standing alone does not have an important bearing on the functional capability of the piping system.
Although individual supports, when considered by themselves may appear to be unstable, it ir necessary only that the entire piping system and associated supports be stable when considered as a single mechanical system.
Dr. Chen describes this system stability as "overall stability," at pages 27-28 of the SIT report.
f Mr. Doyle also agrees that the important criterion by which stability should be assessed is system stability.
Mr. Doyle testified as follows in September, 1982, with respect to a particular support which he observed in performing the STRUDL analyses and which he believed to be unstable (CASE Exhibit 669 at 210):
I wouldn't have done this. I would have used a clamp because, even though the structure below is apparently unstable, it takes so little to make it stable that a support
n horizontally up and downstream is sufficient because the pipe won't see that much force to keep it stable.
Indeed, Mr. Doyle was referring to "a clamp strut situation" similar to that described above (see Figure 2), when he stated that a support up or downstream could easily provide added stability (see CASE Proposed Findings at III-9).
Q. What is standard industry practice regarding consideration of stability of piping systems?
A. As Dr. Chen stated (SIT Report at 28), "it is not general industry practice to explicitly address the overall stability of piping systems together with their supports in design guidelines." Dr. Chen-is correct that stability of piping systems is not explicitly addressed in piping analyses. However, one must also recognize that it is not-necessary to do so because through the normal design process the piping designers achieve a system which will stay within specified deflection limits and, thus, will be incapable of the type of instability illustrated in Figure 1(c). In addition, if the total support scheme does not provide the proper multidirection support required by the piping configuration, the analyst will be unable to.successfully-run the piping analysis computer program (see Tr. 12025 (Bjorkman testimony)). . In summary,-the piping analyst
. assures the stability of the piping system by limiting deflections, which negates any need to assess. stability.
separately.
I
1 i
i Q. Who is responsible for assuring the stability of pipe supports?
A. The support designer is responsible for assuring the stability of each pipe support as part of the piping system, though the pipe may be relied on to supply the stabilizing effect. As stated earlier, this requirement is specified in l
ASME Code Section NF, Appendix XVII, Section XVII-2221(a),
to which all support design organizations are committed.
ALLEGATIONS RAISED BY CASE Q. Please discuss the allegations raised by CASE concerning support instability.
A. CASE has made allegations related to four different categories of rigid body support instability. I shall address each of these individually.
1 7
J
- e
9-I. Box Frames With Single Struts or Snubbers (See CASE Exhibit 6698, item 4i-41)
I I i
~
( \
< )
Figure 3 CASE has alleged that if a clearance exists between the pipe and box frame, the support is unstable, because with a slight eccentricity in upward load, the frame could rotate around the pipe, and the pipe could move upward with no restraining force from the support. CASE has also alleged that if there is 0" clearance between the frame and the pipe, frictional forces needed to resist rotation of the frame are questionable. Applicants do agree with these narrow assertions. However, we do not agree with the conclusion that these factors would always result in the
. supports being unstable. As Applicants testified (Tr.
4788), this type of frame was the result of modifications made in the field. All frames of this type had been.
- identified prior to or during the as-built verification 4
program. Applicants identified 12 such supports in Unit 1 1
and common areas. Appropriate modifications have been made i to provide additional stability to these structures. I l
1 discuss these modifications in more detail below.
i- O. How were the instability problems associated with this type of support identified? In particular, did Mr. Doyle bring this matter to management's attention, as CASE has claimed?
Applicants discovered this condition themselves and promptly 1
A.
acted to address it. The potential instability was identified by ITT engineers on site in May 1981 (see 1
memorandum CPG #36, dated May :22, 1981 (Attachment-A-1)).
As can be seen from page 2 of this memorandum, 1) the two supports referenced as examples (CC-1-159-010-S43R and CC-1-028-039-S33R) are the same supports listed by Mr. Doyle in CASE Exhibit 669B (Items 4g and 4h, 4o and 4p), and 2) the supports were identified as.potentially unstable no later than May 22, 1981. Mr. Walsh began work at Comanche Peak on June 15, 1981. Mr. Doyle began work on August 17, 1981.
Neither of these individuals could have had any part in. the -
identification of supports as potentially unstable.'
Further, Mr. Walsh and Mr. Doyle workea - for either myself or Mr. . Harrison for the entire time they were at CPSES, and could at any time have come to either of us to discuss their. concerns regarding these supports. There was not one occasion when . Mr. . Walsh or Mr. Doyle even hinted .to
Mr. Harrison or myself that they had any concern regarding stability. In fact, only Mr. Doyle alleges that he spoke to anyone regarding his concern. Specifically, he alleges that he had discussed this matter with Mr. Kerlin. However, as 4
Applicants indicated in their Motion for Reconsideration of the Board's Memorandum and Order (at 20-21), Mr. Kerlin was a co-worker of Mr. Doyle's and had no supervisory responsibilities. Thus, Mr. Doyle took no action to bring his concern to any one whom he could have reasonably expected to take any action. Mr. Walsh has not alleged that he spoke to anyone, let alone a supervisor, regarding instability.
O. What action was taken to resolve the questions posed in Memorandum CPG #36 regarding the potentially unstable supports?
l A. As indicated in memorandum CPG #36, Applicants suspended approval of such supports until written procedures were received from ITT in Providence. The ITT engineers at the site continued to pursue a resolution of the problem (see Request for Information RI-9, dated October 12, 1981 and Memorandum dated April 2, 1982 (Attachments A2 and A3)).
When the question had not been resolved by ITT by September of 1982, TUSI directed that the supports be modified to improve their stability.
Q. How many of these types of supports have been identified?
What modifications have been made to correct any instability problem?
A. Applicants reviewed all supports in Unit 1 and common areas and identified 12 frames that fall in this category. (There i
are about 17,000-18,000 safety-related supports in Unit 1 4
4 and common areas at Comanche Peak.) I spoke with Mr. Abele, who wrote memorandum CPG #36 and who at the time was a !
l member of the field engineering group that had made the
. 4 modifications to the supports that resulted in the concern for their stability. He stated that he had communicated ITT's concerns with these frames to the rest of the field I- engineers at the time he wrote his memorandum. Thus, only a few of this type of support were ever constructed.
The types of modifications that have been used on these 1
l frames are shown in Figure 4. They consist of lugs on the '
i i pipe that are indexed to the frames, U-bolts that have been attached to the frame and snugged around the pipe, or the
- addition of extra struts. Each modification will prevent
~
i rotation of the frame around the pipe and thus remove the mechanism th' rough.which the potential rigid body instability could occur.
4-l
i i
b ,+--
1 l m
f f' A Indexed Lugs , U-Bolt Additional Strut Figure 4 Modifications on nine of these twelve frames for stability purposes were initiated before the end of 1982.
The last modification was initiated on February 23, 1983.
l So, all but one of the modifications were initiated prior to the issuance of the SIT Report on February 15, 1983. Thus Dr. Chen's comment in the SIT Report (page 28) that the above methods were "under consideration" must be read in that light.
Q. Is there any validity to CASE's allegation (see Tr. 4955) that Applicants should have written NCRs regarding these supports?
l 1
A. No. In the first instance, this potential deficiency was ,
identified as part of the normal design review process.
Applicants do not utilize NCRs to document deficiencies identified in that process. Second, irrespective of this fact, the Component Modification Cards (CMCs) that were written which created the potential instability do not constitute an official design change until they have been reviewed and approved by the responsible design organization. As I previously testified, all CMCs creating design changes are subject to review by the responsible l design organizations. (Tr. 4870.71.) Thus, any errors identified in the review of a CMC could not represent a l deficiency in design, since the change initiated by the CMC l does not become part of the design until it has been reviewed and approved. In this instance, the potential
- instability was identified prior to approval of the CMCs, and in fact as part of the reviaw cycle. Thus, the question of whether further documentation of this potential deficiency should have been issued is not relevant.
~.s II. U-bolts, Single Struts, and Thermal Gap (See CASE Exhibit 669B, itera 40) i i I e 4 1 "_
l
?, 1/8" GAP Figure 5 Q. Please describe CASE's allegations regarding this type of support.
A. CASE has alleged that this type of support is unstable because the U-bolt clamp assembly will rotate around the pipe under a cot.pressive (upward) load. I addressed this type of support when questioned by Mr. Walsh in September, 1982, concerning CASE Exhibit 669B, page 4Q (a main steam support) (Tr. 4942). I stated at that time, as follows:
"This support is possibly unstable, yes. A final
-. determination as to the stability of this [ type of] support has not been made." At that time in 1982 there had been ongoing discussions in the Technical Services Group and Field Engineering Group at the Site concerning the stability
~
of these supports. These discussions revolved around the impact of various effects on the supports' stability, which are summarized below. (We have attached a full scale drawing of this support configuration to provide a better understanding of the issues. The 1/8" gap is shown in the drawing.) It should be noted that because the design practice for U-bolts used as clamps on single strut supports is to have the U-bolts cinched down (as is done with a clamp), there was no need to impose any restrictions on design practices pending ultimate resolution of thase questions. Continued discussions to resolve these questions and observations took place over a period of several months and involved all design organizations. The specific questions and observations under review were, as follows:
- 1. The pipe stress problems these supports are associated with are each on a straight run of pipe, about 50 feet long, between a penetration anchor and a moment restraint. Therefore, the vertical struts which are essentially plumb will stay that way.
Looking at CASE Exhibit 669B, page 4Q, the Z movement is listed as -1.18". This is the movement along the axis of.the pipe. The movement of the pipe perpendicular to the axis of the pipe is essentially zero. With no eccentricity, tnere would be no force tending to rotate the U-bolt assembly.
In reality, there will be some small amount of out-
of-plumbness in the strut. However, during the compressive load cycle for this support, the crosspiece cannot slide around the pipe. Since there is a large normal force between the crosspiece and the pipe, sufficient friction force will be developed to resist the sliding. So for the U-bolt assembly to rotate, the crosspiece has to roll to one side on the pipe. Looking at the attached full scale drawing (Attachment B) it can easily be seen that the slightest amount of rolling will bring the side of the U-bolt into contact with the piping, s.
producing a friction plane to resist the rotation.
- 2. The upward load on this support which could cause instability has a duration of approximately 1/20 of a second before it reverses direction. It is not likely, and, in fact, is extremely unlikely that the mechanism required to cause instability could be established (an eccentricity needs to occur causing a horizontal component of load which would cause a rotation of the pipe against the cross beam) due to the inherent inertia of the system, and especially l in light of the condition discussed in item 1.
O. Were these supports modified to improve their stability?
A. It bears repeating that the above questions and observations
~
concerning the stability of these supports were under discussion in the fall of 1982. Nevertheless, in late 1982,
_) E .l_C 4.i.. C J j :1.'-[-[-.m
< . . - .n . ,
Sih 4fh, J.fd 1 -.~
~
~
S .,7 J .".
l .
TUSI and ITT agreed to modify these supports to improve their stability because the modifications were relatively simple and readily accomplished. Indeed, I testified to this fact in May, 1983, as follows (Tr. 6735):
JUDGE BLOCH: ...Is that a generic undertaking, the joints of this configuration will have the U-bolts cinched up?
MR. FINNERAN: Yes, in fact, these particular supports, months ago. I participated in investigating in the field with the supervisor of the ITT Design Review Group, in t'rying to determine what methods we could use to improve the stability of these supports, with the addition of additional -
structural steel or by tightening U-bolts.
Q. How many of these supports have been identified?
A. We have identified 15 of these types of supports in Unit i and common areas. Thirteen of these supports are mainsteam supports. Three of the mainsteam supports were actually modified during initial installation in such a way that the potential instability was removed. These modifications occurred prior to September, 1982. The remaining ten mainsteam supports were modified between January,, 1983 and June, 1983, in accordance with my testimony referenced above. The two non-mainsteam supports were modified in October and December, 1982.
The modifications consisted of snugging the U-bolts or adding supplementary structural steel that would prevent the rotation of the U-bolt clamp assembly.
m l
III. Double Strutted Frames (See CASE Exhibit 669B, item 5b)
O. What were CASE's allegations regarding this type of support?
A. CASE has alleged that this type support is unstable.
Applicants disagree. CASE states that this frame will experience out of plane movement of about 1/8", and it will therefore be unstable. However, close inspection of the actual condition shows that this frame is stable.
%1t*
I
- g3 '.
l l s-r.f, h ^M"I :u e__
gan,y r.L L , ,
[ b l O I
Figure 6 The maximum upward load this support could see is 4675 lbs.
(level C). It is this load which would create the
" instability" if no force reacted to prevent the frame from slipping along the pipe. The average center-to-center pin distance on the two vertical struts is approximately 31.5".
For a 1/8" thermal movement axially along the pipe (the maximum this pipe will encounter), the swing angle
- 3 .
'A ,$ _' .,r.:.
w s-> '..L
- k. -
, - .: - .a < ' - " 4 ' ; :* ' '
'l~ '
s.' Y , -
bibwd 55. <+N
- -- e iW ' '
-+-M' A=' "- # 1 -" " 'l'"'"* "-'- ' ' - '- ' * " ' -
"'"#" * ~ ' " '
experienced by the struts would be 0.23 degrees. This makes the maximum vertical component of the strut load equal 4675 multiplied by (cos .230), or 4674#. The maximum horizontal component would be 4675 multiplied by (sin .230), or 19#.
With such a slight inclination of the compression load, the actual load situation with this frame can be simplified as shown in Fig. 6 (cross section view). The resulting compression load condition between the upper and lower pipes and the upper and lower tube steel members is shown in Fig. .
6b. This frame is stable if the frame will not slide along the pipe. To determine whether the loading condition of this support is such that the frame will slide, one should compare the total frictional force that can develop between the pipe and the frame with the horizontal (axial) force acting to slide the frame on the pipe. Assuming the compressive normal force of the strut is imposed, the total frictional resistance would be (.3)(4674) = 1402 lbs., or 30% (coefficient of friction) of the normal load. Thus, the margia against sliding would be 1402/19 = 73. In fact, for any given inclination of load, the relationships of all the forces will remain the same, no matter what the value of the compression load because both the friction force and the normal forces are directly related to the applied load.
Thus, the safety factor against sliding will remain the same, regardless of load, for a given inclination. Because the margin to sliding will increase as the inclination
'M e.' . ',':., . . : wL& .- i L4SC: % L z RA":- ::=W -- k.SS llf l .y .Y T~ A '
j~ ~ ,
.l l
- L ,. .}n!._ .: . i y f : 1. j _. ~' [.[;L 7 3 ] '.{ * . k './ _ l 7 :;, . H ~ .j
! ..- ll
decreases, and we have used the maximum inclination in our calculation, the margin to sliding calculated above is the minimum that would result. Obviously there is no question about the stability of this frame.
IV. Single Struts or Snubbers with Snug U-bolts (CASE Exhild t 669B, items 13u, v and x.)
Q. Would you please describe CASE's allegations regarding these supports?
A. CASE has alleged that this support is unstable because the snug U-bolt cannot provide sufficient friction to prevent -
rotation around the pipe. Applicants disagree. We have performed extensive tests and analyses to determine the capabilities of this type of support configuration to resist rotation of the pipe. As will be discussed in the Affidavit of Dr. Robert C. Iotti and myself regarding the effects of cinching down of U-bolts, these tests and analyses demonstraue that Applicants' original judgment that these assemblies will function as pipe clamps was correct. In short, Applicants' and Dr. Chen's (see SIT Report, p. 28) original judgment that these assemblies would function effectively as clamps was correct.
Q. How many supports of this type have been identified?
A. There are about 380 of these type supports in Unit 1 and common.
Q. How many potentially unstable supports have been identified in the design review processes at CPSES?
A. Almost all Unit 1 and common pipe supports have been vendor certified to date, and as indicated earlier in this affidavit, there are 12 supports which fall into the category I type, and 15 which fall into the category II type. This a total then of 27 safety-related supports for all Unit 1 and common areas that were potentially unstable.1 QUESTIONS RAISED BY THE BOARD Q. Please respond to the specific questions regarding stability raised by the Board in its Memorandum and Order (Reconsideration Concerning Quality Assurance for Design),
dated February 8, 1984.
A. The Board has posed the following questions related to the stability issue:
(1) "[W]hether the forces and moments indicated by the initial pipe run analysis were met by the pipe design groups at the node points to which these supports were attached."
Consideration of these effects is not the responsibility of the pipe design organizations. The support design groups considered all forces and moments provided by the pipe design organization at each node to which the supports were attached. The support designers designed supports that would sustain all the forces and moments that came from the initial piping analysis, which 1
This figure is consistent with Mr. Finneran's representation to the Board in an affidavit filed June 3, 1983, that only 21 of 13,681 supports certified at that time had been identified as potentially unstable.
__._......___.,.______...____.m...- _ . _ _ _ _ . . . . . .
assumes the supports to be stable. Otherwise the support could not accept its intended loads. Obviously, the support designers believed that the designs were stable, therefore properly accounting for nodal forces and moments. In addition, each of the potentially unstable support configurations generated by the field design groups (14 of the 27 supports) were identified in the normal course of the CMC review process. Also, the question of the stability of the main steam supports was being addressed by representatives of the responsible design organizations during the review cycle.
It is important to note that all of the potentially unstable supports were a result of field modifications, with the exception of the 13 main steam supports discussed previously.
(2) "[W]hether all required static and dynamic forces were considered."
As in the normal pipe support design process, all static and dynamic loads were considered for these supports.
However, the question of whether a designer considered all appropriate forces and moments is an independent matter from the question of rigid body stability. A designer will design a support considering all forces and moments. He will also design the support to be stable, and thus maintain its capacity to react those forces and moments. If a designer does not believe a type of support is stable for
l i
the forces and moments he is designing for he will revise the design accordingly. Thus, whether forces and moments are considered is not dependent on the rigid body stability of the support.
The converse of this question is whether stability is Of dependent on consideration of all forces and moments.
course, if a designer failed to consider certain forces and moments the resulting support design could be unstable (i.e., incapable of supporting the applied load). However, .
this was not the case in any of the supports at issue here.
(3) "[T]he nature of the instability, including the conditions under which it would exist and the likelihood of those conditions occurring."
The nature of the instabilities involved has already been discussed in previous sections of this affidavit (e.g.,
seismic thrusts ccmbined with out of plane movement of the support frames). I also discussed several factors which could reasonably be expected to offset any tendency toward instability. Thus, it may be concluded that the conditions which may cause instability are unlikely to occur, but not so readily quantifiable to warrant further investigation rather than implement simple modifications of the supports to improve stability.
(4) "[T]he extent to which Gibbs & Hill was provided with all the information about the performance of the support that they needed for the purpose of doing a revised pipe run analysis and a local pipe stress analysis."
Gibbs & Hill was provided as-built drawings of each pipe support along with as-built survey information that was marked on the drawings. It was not Gibbs & Hill's responsibility, as a piping analyst, to review these supports for stability. It was the support designer's responsibility to provide the piping analyst with what that analyst requested in the analysis, namely, a stable support capable of taking the loads generated by the analyst and in the direction modeled in the piping analysis. If the support designer has designated a support for a particular location, the piping analyst will assume it is stable and capable of taking the imposed loads. He does review the support to assure it is providing a support in the direction
- he has modeled, but does not review each support in detail
to assure that the type of rigid body _ instability we are discussing here does not. exist.
(5) "[T]he reason that these supports were unstable."
The conditions under which these supports may be unstable are addressed earlier in this affidavit. As already noted, it cannot be said with certainty that any one of the 27 problem supports identified by Applicants would have performed in an unstable manner. Nevertheless, as discussed previously, Applicants modified those supports to improve their stability.
(6) "[H]ow Applicant identified these instabilities and the 4
process by which it resolved (or is resolving) them, including the paper trail of that process."
The answer to this question has already been presented previously in this affidavit (see Sections I and II above).
(7) "[T]he potential safety significance of these deficiencies."
With respect to the box frame type supports on single struts or snubbers and two of the thermal gap supports, the potential instability was caused by field modifications which were detected in design review as being unacceptable and were modified accordingly. These conditions did not go undetected in the normal design process, and therefore, should not be considered to pose any question of cafety. As for the main steam supports, the supports were designed by the original designer with the 1/8" thermal gap. For the reasons discussed in Section II above, Applicants and their design organizations believed that realistically these supports would function in a stable manner. Nonetheless,
, ~1 .
1 discussions were held regarding this matter to address the i
uncertainties regarding the supports' stability.
- Consequently, by mutual agreement between TUSI and ITT, the supports were modified to improve their stability.
3 Further, to quantify the safety significance of the potential instability of the mainsteam supports, Applicants have performed an extremely conservative analysis of the j portion of the mainsteam piping which would be supported by l these (and other) hangers. Applicants performed this
) analysis in spite of the fact that they do not believe the allegedly unstable supports would act in an unstable manner.
The stress problem analyzed consists of approximately j 50 feet of straight -32" pipe and a small- segment of 8" i piping connected to the 32" pipe. The 32" pipe extends between a penetration anchor and a moment restraint. In the analysis of this stress problem Applicants assumed that none of the potentially unstable supports would act to carry seismic load, in either direction (although they would carry j load in the tensile (downward) direction even if'they were unstable in the upward direction). In addition, Applicants i
assumed that the remaining supports on the 32" pipe would also not carry seismic load. The stress problem thus analyzed.is supported only at the containment penetration, the moment restraint, and those supports on the 8" connected piping. Results of this analysis indicate that none of the piping stresses-exceed code allowables. The loads on the
containment penetration and the moment restraint are also
'i within allowables. There were some conditions for the supports on the 8" connecting line where the allowables were exceeded. None of these allowables were exceeded to an extent that yielding or failure of the supports would occur.
This analysis demonstrates that not only would the integrity of the piping be maintained but also its functionality even under the unlikely conditions conservatively assumed for the i
analysis. In sum, there is no safety significance to these allegations even if they are assumed to be valid.
Q. Please respond to the Board's two questions in its December 28, 1983, Memorandum and Order (Quality Assurance for Design) (at 21)), which are: "(1) why design guidelines concerning stability were not necessary, and (2) whether design deficiencies are corrected promptly."
A. As to the first question, it has already been shown that the support designer is required to' maintain support stability under Appendix XVII to the NF Code. This requiramont is sufficient in that ill designers are aware of the applicable code ~ requirements.-
It.is also clear that design guidelines for particular types of support configurations were not necessary. Both types of supports determined to be-potentially unstable were-identified at1various stages of the design process without i specific guidance.
Thus, additional guidelines or criteria were not necessaryJto assure a stable piping system. I I
O With respect to the second guestion, as I previously discussed, Applicants identified these potential design deficiencies in the normal course of the design process.
Thus, Applicants promptly identified the potentially unstable conditions. Further, corrective action regarding these supports was initiated in a timely manner, through the ongoing discussions of the issue and the effective hold on further designs of such supports. 'dhere necessary.
I 1
v -r - p , y - , _ _ . . .
l
- 8. te = ,
phnC. Finherman, Jr. f Subscribed and sworn to before me this 17th day of June, 1984.
DVf Notary Public i
3fg (emphat!OM TTWO NIN
- 4 P
4 i
i
$S USSAU W$i INTER OFFICE CORRESPONDENCE i
- ATTACHMENT A-1 OATE: May 22, 1961
,. TO: Ed Eramian i
FAO.W: Gus Abele i
SUS.tECT: C0%1CFE PEM STEAM ELECTRIC STATION BI-WEEKLY REPORT ENDIEG 5-08-81 CPLI d 16_
TSDRE: (5) ITT Grinnell Engineers are participatirg within the Technical Services ;
(TSCRE) are continuing the usual desing review process. Attached is the status re-ports week ending 5-01-81 and 5-08-81.-
F.H.E.: (1) -ITT Representative is reviewing desigr. and reserving tuilding attach-ment area as per daily activities with the layout group.
(1) Engineer is handiing craft problerns generating CMC's via sign cff cycle.
., (6) ITT Grinnell Enginee'rs are assigned to'the Field Desispf Group are implemen-ting design changes via the blue line cycle.
The remaining (6) Engineers are assigned to various ' task forces.
TASK FORCE AREA's STATUS .
Area's 9 & 10 (Aux bldg F1 Elv. 831'-6")' No furthe'r develop, ents .
Safecuard Corridor - No further developments -
r at Suction Valve Rcom (Aux bldg F1 Elv. 810'-6") tio further defelopmeiits ')
.TECK:ICAL ASSISTA.Ms ~
MQer clarificatico rinuired tor carcain d.1 res'acnses reference kr.ter MCI-1457-a include a& tice.31 %:hnical cloud "y s;o.ieets rat' accressed 'previously.
- , loor ,_Q :, . '.c U a ea" ,f Mr . V.W '?.. s! .i:s .icre f ?. ed me,ncli thical'.y ,
b..:fers, i',c i ne e 2.: : 1. r- 50 calied
- t. 3. 2 g r. . . . p ? r.i . ..-!?.' cia a n.}
fixr fini .n c.t.- t4 at:: mined to oe i .v.?".1 ; art eif th : cle::"ar.c i.. :di :naj be , '
tcKen for calcula-icn.; of 'the Hilti embediicnt detth. " " ". .
- 2) Shear Co.nections c - Pipe sucports that'at' tach' tri face Of cantilever steel ' eam c to the base plate by telding along the longitudinal' a.ds.. For wiiici; steel, at:ach2 seat and weld pattern is cut:ide of the bolt pattern. D]ggS'g gjyj$g on gs$ugption3 "
and techniques to qualify these type of connecticns.. - - - -
M. . . . . , M
- Q- r :. L . :: &: :'*..a . ' . .,
N
- 3) Clearance Tolerance - (For restraint box structures)
Relating to Pt 3 on ICN 415, what is the criteria to verify a support if the In preparation 0" clearance cannot be maintained for the vertical direction.
for the As-Built program these cases should be investigated on site.
4)
Please review field procedures for sizing and welding shims per procedure -
No. CP-CFM-9.10 Section 3.3.9 and Attachment 3,4,5 and determine if accep-table for Grinnell hangers.
- 5) Rigid Plate Analysis - The Site is designing and building pipe support building attachments that utilize some sort of Gusset Plate arrange :ent that tne engineers are utilizing various rigid plate evaluation Thistechniques is a oroject that that icnore prying action and flexible plate assumptions.
P.:!. analysis shculd sericusly undertcke and prepare a program sicilar to tae F'JS H fcrc.at. Ple?.se :such base with P.H. analysh and requ;st that analysis pref ire an official prcposal to the site (note cop" Jay Ryan ind Mike Chacer-lain). The selling points wculd be that -he upcc7ing As-Suiu. program ard any new designs would benefit by using a standard and apprcved method i and ro.sL,ble .
i also thg,gustr" er wishes to caintf r. the work en sit _Las much at a Supports th.it intilize single or double trapeze ass.rmb;ies (i.e. struts or snubbers) and a pipe attachment consisting of a steal box frate to restrain vertical and side lotds. Technical Services require concrete guidelines and evaiuation techniques to detemine the stability of using box frames as pipe attachrents due to the physical f tendency of The the box fra:ne to rotata the pipe. Especially for vertical restraints. soverning variables are numerous: pipe size, movement, existing clearance between pipe and frame, single vs, double trapeze, struts installed on skew, location of adjacent supports, vertical vs. horizontal restraints, and a special but comon case of single or double struts and snubbers from beneath pipe. Samplesi are en-closed. (Mark No.'s CC-1-lS9-010-S43R1, CC-1-028-039-533R). Techr! cal m
5ervices willhold all ennnark using box structures as pipe attachments until yritten orocedure or annenval from Providence is received.
Enclosed ccpy of MX f AF-1-036-018-533R. Please forward to product engineering
! 7) f.y a:preval. Single strut on skew frcm the direction of loading greater than clic.,tMe :iccarce of 5 to 10 degrees.
i
- 8) Encics..4 copy of MK ! Ci-1-319-003-5539 as cce our discussion or!cina! Jerign installed utilizing two bolt plates that frame generates weak axis coment about belt axis. Please verify if design is adequate.
- 9) U~ esolved items cer previcus CPG's are:
ai Use of greater embedment lengths for Hilti Kwik Bolts (Rc! CPG v3) o) Ailcwable stresses for Fig.140 threaded rods utilizc1 as thru-slab appli:ation anchor bolts (REF CPG # 23) c) app
- oval of greater Hilti Shear vi.lves as per (CFG r 30)
Engineeriag Aids Pente! ifecs Pencils (3) Cozen
^'
Hangar Tapiates 21 Dnzen ITT Catalci 1) Cozen Pink Pesrl erasers type 101 3) Dozen N '- 0 ^ ..l.<
I
^
^YO ES 014
.iEQUEST FOR INFORMATION
[* COMANCHE PEAX SITE #9 U ~~ f Tom a MANAGER PRODUCT ENGINEERING a MANAGER PIPING & STRUC. ANALYSIS O DIVISION Q.A. MANAGER j
=a MANAGER R0hE enhart Mul.] 'rHIS COPY FO C MANAGER APPLICATION ENGINEERING l i
O OTHER 4
l FRON= nne as.1._1.mA 3 4+= [-,a* , Comanche Peak DATES in-12 n1
)
INFORMATION REQUESTEDs suonorts that utilire sinale ac dnuble tranere etsel meesenhlf ac fi e strute ne ennhharel and a nina attmehment enneletinn af a hnr frams tn restrain vertical and tid. Inade fTT Grinnell field enaineert rennfre ennerste nuidelinee and mumluatinn technfanne tn Amtsegn4ne the etah414tv af
..einn haw fe.=me me n4na atturhmanee A,im ta +hm nhve t e n1 +mnAmarv af the haw #,mma ta entats arannd the nine_ senarially far vertical reetrainte The navsenina varishlee are namnsenuta nina etwo mnvasnant avietinn elemens ca hahasan nins and frame _ cinnie we harirnntal restrainte . and a ere ial hut ennunna emes af ef_nnis I
3r double struts and snubbers from beneath the ofne (see attachments) i i
I i
Dav* Me"5 ,
! CCs CD MANAGER APPLICATION ENGINEERING i
CD MANAGER SITE ENGINEEING rd o-*<3FA E-i i -
O MANAGER SPECIAL PROJECTS O MANAGER R0hE 1
O OTHER ACKNOW_m*hf.NTe THE A80VE INFORMATION i HAS BEEN RECEIVED, RESPONSE TO THIS l
REQUEST W!t.L BE ISSUED BY l
DATEe I ACMNOWLEDGED SYe
.w w ww . . . . .
c . A;tM m. . .. . ... . .. , . . :
.r . . . a. . :
m Lu m~_
DATE . I . .* SUBJECT. %EET NO. . .. . OF k.
?. .
(f. ..
. DATE . . CUSTCuER . . . ... M- . . .
CHED. gy , ,
I PROJECT..... . . . . . . . . . . . . . . . . . PROJECT NO. . . . . .......... .... ...
l
\. N b A
$$'fh. . -
0 !! .
s:.@f -
Y _p _ v 4:.sydT
_g _
A
- h) . .
_r
_ E EVA 7ck... Vl EW. _.. .. .-... . . .
O Wi. .. , -
r,. ,
.s'h l
l F- =::
l
_ 9NPQ.7T7 l
M -
-tf 4 A>
-wl -
T~~~
7EEVA1od - ~ vie _d I
I Medy w-- m=o@.%31op-..
r, doMAMCWE PEKStit: "" @ 97
... .x . ...,,..
. ~ m. .
.. 's _n _. m. .
._. e x.x cp3_. ,L . .. . . . . . ,. .
ATTACIIMENT A-3 TO: GUS ABELE Eh* TE: 4;gjeg FAOMt RONWISNIEWSKI[5)
SUS /ECT: REQUEST FOR INFORMATION DATED 10/12/81 (RI-9) 4 A general answer cannot be generated per the above request for the following reasons:
- 1. Hcw can the examples shown restrain side loads?
- 2. Could you define the stability problem in greater detail?
- 3. What is the weld configuration?
- 4. Do you have 2 or 3 assemblies of this type or 20007 If you only have a few, why don't you send -
them over via an SA request?
I would suggest discussing problems of this. type with Ed Eramian first or give me a call on Extension 422 and I will be happy to discuss them with you. Please reference RI-9 in all future correspondence concerning the above request.
RW/msb cc: E. Eracian D. Powers F. Vasilladis 4
_ .. a . . . . .. . .t w .. ., s . . . o .- > .- -
h
. O l l l l -
WDOCUMENT
. ~
l -
~'
. PAGE~ .
PULLED
~
l i
l ANO~ eneesa .
l NO. OF PAGES __ \
REASON l ,
D PAGE ILLEGS1 CF l
. D MAC CCPIFa!D A1. o,e CD- -
J >_ l I
D BETE = CCP(REOJESTED ON _
---g nan:>cce< rad At en @
D b
. , ci,<re b N d 3 2-dl D FAMED ON APER 1gRE CNt> NO
--r-,-