ML20137C447
| ML20137C447 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 06/07/1985 |
| From: | Chen W NRC - COMANCHE PEAK PROJECT (TECHNICAL REVIEW TEAM) |
| To: | |
| Shared Package | |
| ML20136F556 | List: |
| References | |
| NUDOCS 8508220267 | |
| Download: ML20137C447 (9) | |
Text
.
Disk: MISC 2 Job: u-bolt ENCLOSURE CPSES - ASSESSMENT OF SIGNIFICANCE OF APPLICANTS' U-BOLT SAMPLE DISCUSSED IN APPLICANT'S MOTION FOR
SUMMARY
DISPOSITION REGARDING CINCHED DOWN U-BOLTS INTRODUCTION A review was conducted to detennine whether the U-bolt sample discussed in Applicant's motion for sumary disposition and associated affidavit and statement of material facts relating to CASE's allegations regarding cinching down of U-bolts in pipe supports (References 1, 2 and 3) was
" randomly selected" and " representative", as characterized in Applicants' Reference 2 affidavit, and the relevance of the accuracy of Applicants' characterization of their U-bolt sample to the technical issues in Applicants' motion for sumary disposition. The review was based on Applicants' motion and other related documents (listed in the References section) as well as information provided by Applicants in an April 22, 1985 meeting with the Staff (Reference 4).
The review determined that Applicants' U-bolt sample was not randomly selected in a statistical sense, but appears to be reasonably represen-tative. Provided Applicants fulfill their comitment to conduct a 100 percent inspection of U-bolt supports during which the U-bolts are torqued to values satisfactory to the Staff, the question of whether or not Applicants' U-bolt sample was randomly selected and representative is not technically important from the standpoint of safety with respect to stability of the pipe supports and stresses in the U-bolts and supported piping.
It is not clear whether the characterization of the the sample as random and representative is important from the standpoint of demonstrating the adequacy of Applicants' design and construction practices, since the basis for Applicants' assertions concerning these practices is not clear in the Motion.
BACKGROUND Applicants' motion for sumary disposition was submitted as part of their plan (References 5 and 6 to respond to the CPSES ASLB design QA memoranda (References 7 and 8))regarding the adequacy of the design of pipe supports in the CPSES plant. Specifically, Applicants' motion for sumary disposition was intended to provide evidence that their design and construction practices for cinched down U-bolts in snubber and strut type supports were adequate to provide sufficient stability, and did not result in unacceptable stresses in either the U-bolt or the piping. This evidence was to be obtained by a scven part test and analysis program (Reference 1, p. 4 and pp. 5-10; Reference 2, p. 4 and pp. 12-33; Reference 3, pp. 2-7 and Reference 4, pp. 16-22). Prior to performing these tests and analyses, Applicants conducted a field inspection of U-bolt torques (Reference 2, pp. 9-10; see also Reference 1, p. 5; Reference 3, p. 2 and Reference 4, pp. f2 2W Applicants explained the need for the sampling program as follows:
850022o267ByBd0gs PDR ADOCK PDR A
.' Q.
In that the objectives of the tests [and corresponding finite element analyses] rely on the preload (torques applied-to the l
[U-] bolts) in the field, how did you detennine the range of preloads to use?
A.
Until now, the Board's sole information regarding the preload level that would exist in the field was based on the Brown and Root Design Change Notice No. 1, dated October 8, 1982, to Construction Procedure No. 35-1195-CPM 9.10 Rev. 8...
This [ procedure] has been interpreted by Mr. Doyle as resulting in approximately 800 in-lbs. of torque... Mr. Doyle made no attempt to quantify whether this value of torque is applicable to all pipe sizes, and all pipe schedules. Applicants have done so.
To determine the range of torques which exists in the field Applicants inspected the torque of a randomly selected represen-tative sample of cinched down U-bolt supports. This data was used to determine the range of torque to be applied to each of the test specimens.
The results of Applicants' survey were summarized in Table 2 of their affidavit, which listed " torque ranges" for 160 U-bolts in supports for 17 pipe sizes varying between 3/4 inch and 40 inch (Reference 2, Table 2).
According to Applicants, their survey data established (Reference 2,
- p. 10; see also Reference 1, p(. 5; Reference 3, p. 2 and Reference 4, pp. 31-3BT that in their test and consequently, their associated analysis) program that:
[F]or the four inch pipe, tests should be conducted with preload varying from 5 to 60 ft-lbs. Similarly, a torque range of 10 to 100 ft-lbs. was established for the 10-inch pipe tests, and a 20 to 240 ft-lbs. range was established for the 32-inch pipe test.
i' Applicants stated (Reference 1, p. 5) with respect to CASE's concerns that their test and analysis program showed that:
i
[N]one of CASE's... concerns raises an issue that reflects a breakdown in Applicants' Quality Assurance ("QA) Program or a safety concern in the plant.
Applicants also claimed (Reference 2, p. 34; see also Reference 1, p.10; Reference 3, p. 27 and Reference 4, pp. 27 7ET with respect to stability that their test program had concluded that:
l
[T]he U-bolt / cross piece assembly can perfonn effectively as. a clamp provided that sufficient preload is established in the U-bolt (it should be noted that a clamp also requires preloading). -We further conclude that even if the preload was insufficient, but still present in some amount, the U-bolt support would vibrate, but still be capable of supporting the necessary load, thus behaving
" stably."
Nonetheless, Applicants indicated that to provide further assugnce that the preload on all affected cinched down U-bolts is adequate (Reference 2, p. 34; see also Reference 1, p. 11; Reference 3, pp. 7-8 and Reference 4, pp. Hi 30 and p. 40):
Applicants will conduct a 100 percent inspection of the torque of all such U-bolts (380). At the time of inspection, to remove questions regarding stability, Applicants will assure that such U-bolts are torqued to levels at which the [U-bolt] assemblies will be stable in the absolute truest sense, i.e. no rotation, and axial motion, if any, is toward the strut.
Torque values for Applicants' 100 percent U-bolt inspection program were set forth in Table P of their U-bolt affidavit for the four pipe sizes considered in Applicants' test and analysis program (Reference 2, p. 74 Table-P); torque values were to be established by Applicants for other pipe sizes (Reference 2, p. 74). Applicants further stated that their analysis program showed the U-bolt supports would be stable at the minimum torque values considered in their analysis (Reference 2, p. 72),
which were' smaller than the corresponding torque values to be utilized in Applicants 100 percent inspection program.
During the August 6,1984 discussions between Applicants, CASE and the Staff on Applicants' motions for sumary dispositions, CASE witness Mark Walsh raised the following) questions relating to Applicants' U-bolt sample (Reference 9, p.~ 11 :
We would like to know what criteria was used to select the particular supports, how many of these supports have the U-boltscinchdownafterthe[BrownandRoot]procedurebefore l
this cinching of U-bolts came into effect, were in the sample.
l How the random sample was selected, and how the random sample l
is representative with the bolts, all the U-bolts in the plant.
t.
-1/
Applicants had noted in their accelerated vibration test that (Reference 2, p. 27; see also Reference 4, p. 27 and p. 40):
Insufficient [U-bolt) assembly preload will permit the assembly to rotate about the pipe and also to walk axially along the pipe axis in either direction with respect to the location of l
the strut.
4-Applicants responded by stating (Reference 9, p. 12):
[T]he sample was random in the sense that we did not choose any particular type of U-bolt. We chose reasonably, with a criterion that said we should C-bolt, not U-bolt in different sizes, in random areas of the plant.
Subsequently, Applicants identified the supports in their U-bolt sample and provided data relating to the torques in the legs of the U-bolts in their sample (References 10,11 and 12) in response to Board memoranda requesting the raw data on U-bolt torques in Applicants' sample (References 13 and 14). Based on Applicants' Reference 10 response, CASE moved that the Board find that Applicants had made material false state-ments to the Board (Reference 15 and 16). Applicants responded to CASE's motions in References 17 and 18. Subsequently, the Board ruled that Applicants had made misleading statements and reopened discovery (Refer-ence 19). Applicants subsequently moved that the Board reconsider its decision (References 20 and 21).
DISCUSSION A comparison of the U-bolt torque data for Applicants' U-bolt support sample in Table 2 of Applicants' U-bolt affidavit (Reference 2) with data provided by Applicants in References 10, 11 and 12 found that:
1.
All 160 supports in Applicants' sample were Unit 2 supports.
2.
Of the 160 supports in Applicants' sample, 52 were non-safety related supports.
3.
Of the 160 supports in Applicants' sample, 36 were for pipe sizes 3 inches and under.
4.
The ratio of the torques in the legs of the U-bolts could be as high as 3 but generally were between 1.0 and 1.5.
5.
The torque ranges in Table 2 of Applicants' affidavit (Reference 2) were based on the average of the torques in the 2 legs of the U-bolts.
Random Sample With respect to Applicants' characterization that their U-bolt sample was " randomly selected", based on information provided by Applicants in the April 22, 1985 meeting (Reference 4) and References 12 and 20, it was determined that the U-bolt sample was not randomly selected from Units 1 and 2 and Comon Areas and might be biased towards readily accessible U-bolts in Unit 2 supports. First, Applicants did not perform
" detailed statistical evaluation and analysis" of their sample (Reference 20,
- p. 8). Second, although Applicants state that their sample was selected "at random" without "any intentional bias" from unpainted cinched down U-bolt supports that could be found in Unit 2 during essentially 2 days k
. ofsamplingj(Reference 4,p.15andpp.58-62;seealsoReference12, pp. 2-3), - the raw data in Reference 6 suggests that the bulk of the sample were accessible.ptainedfromUnit2supportswhichmighthavebeeneasily l
Representative Sample With respect to Applicants' characterization that their U-bolt sample was representative, the review found that although the sample was not representetive of Units 1 and 2 and Comon Areas, the sample was sufficiently varied to provide an indication of the range of torques in U-bolts installed in Unit 2 and could be characterized as representative of the U-bolts installed in Units 1 and 2 and comon areas.
First, a review of the data in References 10 and 11 indicates that U-bolts are found in supports for various pipe sizes throughout the Unit 2 reactor, safeguard and turbine building in varying piping systems including the ContainmentSpray(CT),ComponentCooling(CC),SafetyInjection(SI),
Main Steam (MS) and Feedwater (FW) Systems. Second, Applicants explain that their sample was limited to unpainted Unit 2 U-bolt supports since the supports in Unit I and Comon Areas were painted and torque data from painted supports would have been invalid; the torque readings would be greater than the actual U-bolt torques due to the shear resistance of ~
the paint (Reference 4, pp. 52-54; see also Reference 12, pp. 2-3; Reference 17, pp. 3-5 and Reference Ei, pp. 9-10). Third Applicants state that the Unit 2 U-bolts (both safety related and non-safety related) sampled were identical in make, manufacture and sizes as the U-bolts in Unit 1 and Comon Areas (Reference 17, p. 4 and Reference 18, pp. 1-2).
Fourth, Applicants claim that ti.e October 8, 1982 design change notice to their construction procedure did not result in any change in their construction practice relating to U-bolt torquing; the 2/
Applicants stated, with respect to the sampling program, that no formal procedures were written for the survey. The 3 Pipe Support Engineering engineers who collected the data were instructed verbally to measure and record the torques on both nuts on any cinched down U-bolt that could be found in Unit 2 that were unpainted. The torque in both legs of the U-bolts were to be measured where accessible. Safety related and non-safety related supports as well as supports installed prior to and subsequent to the October 1982 change notice to Applicants construction procedure were included in the sample.
3/
Applicants stated that most of the sampling was conducted on two days. A review of the raw data in Reference 6 indicated, however, that approximately 2/3 of the sample (105 of which 94 were reported) was obtained on January 21, 1984 and approximately 1/3 (45 of which 40 were reported) was obtained on January 28, 1984 Only 16 supports were sampled between April 14, 1984 and May 24, 1984.
change notice documented their construction practice and was written at the request of the NRC resident inspector. Moreover, Applicants state that their review of torque data in their Unit 2 U-bolt sample indicated that the U-bolt torques appeared to be unaffected by the October 8,1982 design change notice to their construction procedure; the distribution of i:I torques in the Unit 2 supports in their sample installed prior to the design change notice and in all the Unit 2 supports in their sample appear to be similar (Reference 4, pp. 54-56 and pp. 63-67; see also Reference 17, p. 7; Reference 18, pp. 1-2; Reference 20, pp. TT-IY and Reference 21, pp. 3-4).
Fifth, Applicants claim that their construction practice for torquing U-bolts in Unit 1. Unit 2 and Common Areas was the same. Applicants state that they identified 45 construction foremen whose crews installed 122 of the 160 U-bolts in their sample and that their interviews with 28 of the 45 foremen who were still on site deter-mined that 25 of the 28 foremen and their installation crews had worked in both Unit 1 and Unit 2.
Applicants noted that there was an overlap in construction activities in Unit I and Unit 2 between late 1977 and late 1983 (Reference 4, pp. 60-67; see also Reference 17, p. 6; Reference 18, pp. 1-2; Reference 20, p. 11 and Reference 21, pp. 3-4).
Finally, Appli-cants claim that their construction practice for torquing U-bolts in both safety related and non-safety related supports were the same, based upon
' heir comparison pf torques measured in safety related and non-safety related U-bolts 4 (Reference 17, pp. 6-7 and Reference 18, pp. 2-3).
Based on the qualitative information outlined above, it appears that data from Applicants' Unit 2 U-bolt sample can be characterized as "representa-tive" of U-bolts installed in Units 1 and 2 and Common Areas. It should be recognized that this conclusion is a relatively qualitative judgment, and that a quantitative statistical evaluation and analysis of the size and composition of Applicants' sample in support of this conclusion has not been performed, either by Applicants or the Staff.
Technical Significance The accuracy of Applicants' characterization of their field survey as
" random" and " representative" is not technically important from the j
standpoint of safety wth respect to stability of U-bolt supports and l
stresses in the U-bolts and the supported piping. Applicants committed to conduct a 100 percent inspection of the U-bolts in the plant during i
l which the U-bolts will be torqued to levels determined by Applicants to be high enough to assure that the U-bolt assemblies will not rotate around the pipe nor walk axially along the pipe while transmitting their design loads (characterized by Applicants in Reference 2, p. 40 as l
" stable in the absolute truest sense"), yet low enough to assure that stresses in the U-bolt and supported pipe will be acceptable. However, l
Staff has not confirmed that the torque values to be used in Applicants' l
100 percent inspection will be acceptable; technical issues relating to l
Applicants' test and analysis program described in Applicants' motion l
\\
4/
Applicants have not provided such a comparison to date.
4.
have not been resolved. Provided that these technical issues are satis-factorily resolved and hence the torque values to be used in the 100 percent inspection are satisfactory to the Staff, the retorquing'will assure that U-bolts in the facility provide adequate stability without excessive stress.
i With respect to the issue of whether or not Applicants' design and construction practices for cinched down U-bolts provided adequate stability and did not result in unacceptable stresses, it is unclear that the characterization of the sample as " random" and " representative":
affects this, since the basis for Applicants' assertions are not clear.
The evidence in the motion indicates the need for modification of Appli-cants' construction practice with respect to cinching down U-bolts.
Applicants did not discuss this implication; instead they committed to retorque the U-bolts to provide "further assurance".
O w&,,-
-_m
1,
REFERENCES 1.
Applicant's Motion for Summary Disposition of CASE's Allegat' ions Regarding Cinching Down of U-bolts, June 29, 1984.
2.
Affidavit of Robert C. Iotti and John C. Finneran Jr. Regarding Cinching Down of U-Bolts, June 29, 1984.
3.
Applicants' Statement of Material Facts As To Which There Is No Genuine Issue Regarding Consideration of Cinching U-Bolts, June 29, 1984.
4.
Technical Conference between Texas Utilities and Comanche Peak Project Team, re: Applicants' U-bolt sample, whether it is repre-sentative of Unit 1 April 22, 1985.
5.
Applicants' Plan to Respond to Memorandum and Order (Quality Assurance of Design) February 3,1984.
6.
Supplement to Applicants' Plan to Respond to Memorandum and Order (Quality Assurance for Design), March 13, 1984.
7.
ASLB Memorandum and Order (Quality Assurance for Design),
December 28, 1983.
8.
ASLBMemorandumandOrder(ReconsiderationConcerningQuality Assurance for Design), February 8, 1984.
9.
Discussions on Motions for Summary Disposition Filed by Applicant, Comanche Peak, August 6, 1984.
- 10. Applicants' Response to Board Request for Information Regarding Cinching Down U-Bolts. October 23, 1984.
- 11. Applicant's Response to Board Request for Raw Data Regarding l
Cinching Down U-Bolts, November 9,1984.
- 12. Affidavit of John C. Finneran Jr. Regarding Information Related to Cinching Down of U-Bolts, November 9,1984.
- 14. ASLBMemorandumandOrder(RawDataonU-Bolts). October 24, 1984.
- 15. CASE's Motions and CASE's Answer to Applicant's Response to Board Request for Information Regarding Cinching Down U-Bolts, November 5, 1904.
l l
, 16. CASE's Answer to Applicants' Response to Board Request for -
Information Regarding Cinching Down U-Bolts in the Form of Affidavit of CASE Witness Jack Doyle, November 5, 1984.
- 17. Applicants' Reply to CASE's Motion Concerning Information Regarding Cinching Down U-Bolts, November 19, 1984.
- 18. Affidavit of John C. Finneran Jr. Regarding CASE's Motion Concerning Sampling of Cinched Down U-Bolts November 19, 1984.
- 19. ASLB Memorandum and Order (Reopening Discovery, Misleading Statement) ASLBP No. 79-430-06-OL, December 18, 1984.
1.
- 20. TUEC Motion for Reconsideration of Licensing Boards' Memorandum (Reopering Discovery; Misleading Statement, January 7, 1985 (misdated January 7, 1984).
_,, _...,