ML20100K271
| ML20100K271 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 11/06/1984 |
| From: | Williams N CYGNA ENERGY SERVICES |
| To: | Burwell S NRC |
| References | |
| 83090.021, NUDOCS 8412110032 | |
| Download: ML20100K271 (15) | |
Text
- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
y ,
e 101 Cahfornia Street. Suite 1000, San Francisco, CA 941115894 415/397-5600 Novenber 6,1984 83090.021 Mr. S. Burwell Licensing Project Manager U. S. Nuclear Regulatory Commission 7920 Norfolk Avenue Bethesda, Maryland 20814 Subj ect: Response to NRC Questions Comanche Peak Steam Electric Station Independent Assessnent Program - Phases 1 and 2 Texas Utilities Generating Company Job No. 83090
References:
(1) NRC letter of March 30, 1984 from D.G. Eisenhut to M.D. Spence and L.L. Kammerzell.
(2) Minutes of meetir.g between NRC/TUGC9/Cygna on April 19, 1984, pgs. 87-91 and 160-162.
(3) Minutes of meeting between NRC/Cygna on July 3,1984, pgs.
43-49 and 50-55
Dear Mr. Burwell:
In response to the attached telecon between N. Williams (Cygna) and D. Wade (TUGCO), Cygna has prepared the following responses to the NRC questions which remain open at this time. Questions associated with cable tray supports are not provided in this letter since those reviews are still ongoing.
16tC Question #1 Is it appropriate for hardware to be installed in a manner different from that shown on the design drawing if this is allowed by a procedure? (Reference 3, pg. 44).
Cygna Response Cygna has reviewed the Phase 1, 2 and 3 assessment scopes and has determined that the only example of this situation is the practice of reversir.g snubbers 180* from the design drawing orientation.
In this case, a procedure was used to modify a design drawing without a design change or nonconformance report (NCR). It is Cygna s opinion that the design drawing takes precedence over the procedure and that changes to the drawing, ,
i.e., the end to end reversal, would require a drawing revision including the 1 complete drawing review and approval process. Alternately, an NCR could be ,
issued and dispositioned " accept as is" by TUGC0 engineering with the NCR l
s -- -,~ - -
8412110032 841106 n.- -., , s -- - gzz , ,
L PDR ADOCK 00000445 i'$
A__-_________--_ CDR i --
r
\
Mr. S. Burwell November 6, 1984 Page 2 1
attached to the as-built drawing. Since the installation procedure (CP-CPM 9.17 Revision 2) applies to snubbers on a generic basis, snubbers which are reversed must be verified to be acceptable by TUGC0 engineering via a controlled design l change method such as a drawing revision, design change notice or NCR. TUGC0 engineering did not approve procedure CP-CPM 9.17 Rev. 2, therefore, the j existing condition violates the following requirement of 10CFR50 Appendix B, Criterion III:
Design changes, including field changes, shall be subject to design control measures commensurate with those i applied to the original design and shall be approved by the organization that performed the original design unless the applicant designates another responsible organization.
Cygna would like to point out that this example has no design impact at CPSES. It is primarily a question of accepting a design deviation without appropriate engineering approvals. If additional documentation exists which l demonstrates that Criterion III has been satisfied, then Cygna concurs with '
TUGCO's use of installation procedures to allow for the end for end reversal.
NRC (Nestions #2 If certain support designs at CPSES deviate from standard industry designs, is it valid to use standard engineering procedures to design them? (Reference 2, pg. 160.)
Cygna Response l l
Some possible examples of nonstandard designs discovered during the Cygna reviews are:
- 1. The consistent use of tube steel in pipe support designs without .
providing local stress checks when appropriate. l
- 2. The practice of sometimes using trunnions in place of shear lugs.
- 3. The practice of neglecting self weight excitation for relatively large frames.
4 Use of Richmond Inserts through tube steel for attachment details.
- 5. Cinching of standard U-Bolts to perfonn the function of a clamp.
l l
l I
r s scss _ j Mr. S. Burwell November 6, 1984 Page 3
- 6. Use of box frames (with 0" gap) as pipe clamps.
Cygna has attempted to address this question during the course of the Phase 3 reviews. To document the results of this review, a set of General Notes to the Phase 3 pipe support checklists was prepared and included in Volume 2 of Final Report TR-84042-01. A copy of these notes is attached for your con-venience. Each apparently " nonstandard" design at CPSES was reviewed by Cygna for acceptability and any generic implications. The corresponding general note, observation, or report section which addresses each of the above examples is as follows:
Example 1 - Pipe Support Checklist General Note 3 Example 2 - Pipe Support Checklist General Note 2, Observation PI-06-01 and Section S.1 of the text Example 3 - Pipe Support Checklist General Note 7 Example 4 - Pipe Support Checklist General Note 6 Example 5 - Pipe Support Checklist General Note 12 Example 6 - Pipe Support Checklist General Note 16 Example 5, cinching of U-bolts, is still open pending acceptance of the U-bolt testing and analysis program recently completed by TUGCO. While these appli-cations are not necessarily standard compared to those suggested by the Codes or used by other design organizations, TUGC0 has performed analyses to show that these designs do meet the stress allowables set forth by the ASME Code, Subsection NF. Cygna's detailed review shows these designs do meet the Code allowables and are adequate to perform their intended function.
If you wish to discuss these matters in more detail, please feel free to call at your convenience.
Very truly yours, k .MI%
N. H. Williams Proj ect Manager dmm Attachments cc: Mr. J. B. George Mr. D. H. Wade Mr. D. Pigott Mr. S. Treby Mrs. J. Ellis
Communications 4L t i Report llllllllllilllll111llllllllll1 Texas Utilities T*' econ g conference neport Project: Job No.
83090 Comanche Peak Steam Electric Station U*
Independent Assessment Program - Phases 1 and 2 10/12/84
Subject:
Time:
Phases 1 and 2 Open Items List from NRC 3: 00 P.M.
Place:
Participants:
of N. Williams Cygna nequired item Comments Action oy S. Burwell (NRC) had telephoned D. Wade to relay a list of open items remaining for close-out of the Phase 1 and 2 reviews. D.
Wade relayed these items to N. Williams for action by Cygna:
- 1. With information available from the Phase 4 cable tray walkdowns, can Cygna issue a letter stating whether or not the cable trays have been constructed in accordance with the latest drawings?
- 2. Referencing the April 19 (pp. 160-162) and July 3 (pp. 50-65)
NRC/Cygna meeting transcripts, the NRC would like Cygna's assessment of whether any problems have been identified due to the use of "non-standard" pipe support designs. For example, the effect of "non-standard" supports on the design assumptions and procedures.
- 3. Regarding the July 3 NRC/Cygna meeting where the scope of Cygna's welded attachment review was discussed, the NRC understands that further work is still being performed. N. Williams replied that double trunnions used as axial and/or rotational restraints was the only action item still open from Phase 3. Further, J.
Minichiello (Cygna) had a discussion with D. Terao (NRC) on September 21, 1984, where this was explained. D. Wade will contact the NRC again to determine whether the work performed to date is sufficient to address the NRC concerns with the use of welded attachments at Comanche Peak.
D " b u'
- a YhtbblVik ~
/alb ~ 1 N. Williams, D. Wade. J. Van Amerongen, J. Minichiallo, L. Weingart, G. Bj orkman, 2
i n..., o u_ m. e ._ n ,1 _ _ _ _ _ .. m _,, ,
, . -y 7 ., ,. ., .w . ,,.,,,,o,y, v. Lii sa, rsvjeLL
Communications A( t i Report 11111llllllll11141111111111111 nem comments $:Y/n'Iy 4 With the information available from the Phase 4 cable tray support reviews, can Cygna issue a letter stating whether or not the design drawings match the construction drawings?
- 5. Referencing the April 19 (p. 91) and July 3 (pp. 44-49) NRC/Cygna meeting transcripts, the NRC would like Cygna's opinion on the appropriateness of maintaining information in procedures which allows for installation of pipe supports in a manner different than that shown on the drawing.
N. Williams will review these questions with the appropriate Cygna reviewers and provide those responses which Cygna has sufficient information available at this time to prepare a response as soon as possible. Cygna has not yet completed those-aspects of the cable tray support reviews necessary to respond fully to the NRC questions. N. Williams agreed to explore the possibility of extracting the necessary information, but it did not appear promising.
Page of tom a te
GENERAL IWTES TO PIPE SUPPORT CHECKLISTS
- 1. Component Weights As a matter of standard practice, the pipe support design organizations do not include standard component weights (i.e., strut, spring, snubber, clamp) as part of the pipe support design load. They normally consider the weight of the frame members when using the STRUDL program for design, however, they neglect the standard component effect. Since these components weights are typically small in comparison to the applied ,
pipe load (5% or less), they will have little impact on design, even in the case of the weight being orthogonal to the applied load. In addition, it is common practice to neglect these weights for struts, snubbers, and rods. Cygna has seen examples in industry where the weight of large constant supports is .
included in the design of the wall or ceiling attachments but these are typically no more than 5% of the pipe load and can be considered negligible. Therefore, Cygna finds this procedure acceptable.
- 2. Pad / Trunnion Stresses on the Main Steam Line In the pipe support calculations involving pads or trunnions welded to the Main Steam piping, Cygna did not find many examples of stress checks. Instead, the drawings carried the note " Pad (or trunnion) qualified per Apperdix G of ASME B&PV Code." Per TUGC0 document CPP 12978, attachments welded to the Main Steam and Feedwater lines require impact testing (per Subsection NC-2311 of the ASME B&PV Code) or assurance that the stress levels are low enough to preclude non-ductile failure. In order to qualify pads or trunnions already assembled, NPSI (Secaucus) performed detailed finite element analyses of each geometry and compared the maximum stresses to allowables derived from Appendix G (Prevention Against Non-Ductile Failure), which resulted in stresses much lower than standard Code allowables. Cygna reviewed two examples of the NPSI .
models/ calculations and found their method acceptable, although one model contained input errors which did not impact the conclusions. Furthermore, as part of their normal design practice NPSI had previously committed to reviewing each welded attachment analysis against the final pipe support loads (refer to Communications Report dated 6/18/84). Thus, Cygna con'siders the approach acceptable.
- 3. Local Stress Effects ,
In reviewing the pipe supports for Phase 3, Cygra noted many instances of the following:
a) lise of wide flange or back to back channels without stiffener plates at connections and without calculations to show the joint is acceptable.
r _ yp
=Jk ie Texas Utilities Electric Company; Independent Assessment Program, Phase 3 EME#EEE Final Report TR-84042-01, Rev. 1 Page 1 of 9
b) Use of tubesteel 'in frames without checking whether the webs of the tube are adequate to transmit the load, especially when the end is a load transfer point.
c) Use of composite sections, made up by welding a plate to a tube section, without considering the additional stress in the weld at the load transfer point (see Observation PS-07).
It is important to note that Cygna did see instances where each of these items were properly considered, either by calculation or good design practice. In response to Cygna's question on project guidance in this area, TUGC0 stated in a 6/8/84 letter:
"Although the various design guidelines may not require that specific calculations be performed .
on structural connections the effects of localized stress am often evaluated with approximate calculations. The individual design engineer assesses each situation on a case-by-case basis.
From his inspection, he may judge the effects negligible or may add gussets or stiffeners; or he may elect to calculate the actual stresses and determine if there is a necessity for stiffening. In all cases, however, the designer is guided by the limits set forth in subsection NF and specification MS-46A. It has always been a matter of good engineering practice to make these considerations. It is not industry practice to provide guidelines to engineers for these considerations, nor is it necessary."
Cygna has reviewed each design in Phase 3 for the acceptability of the engineering judgment noted. In certain cases, Cygna was able to confirm that judgment since the applied loads were small. In other cases, Cygna performed their own calculations to determine the adequacy of the joint. In no case did Cygna find a design error, i.e., each joint would transfer the applied loads. It is the lack of calculations or notes in the design calculations that has caused Cygna to make this comment. Without at least a statement such as: " connections OK by judgment", Cygna had no way of knowing whether certain joints had been checked or not. Conversely, if stiffeners were added to a joint without calculations, Cygna had no means of determining that the stiffeners were properly designed, without performing our own calculations. Thus, while the lack of calculations in this area made the review more difficult, Cygna did not find any instances of overstress due to inadequate engineering judgment, r y, k bi#
'4 I Texas Utilities Electric Company;
= '
Independent Assessment Program, Phase 3 MMMMNMMMM Final Report TR-84042-01, Rev.1 Page 2 of 9
- 4. 3-Sided Welds In certain connections, Cygna noted the use of 3-sided (see sketch) welds used to transfer the loads from one member to another. In most instances, the designer did not transfer the loads from the center of gravity of the beam (Point A) to the center of gravity of the weld (Point B). It is TUGCO's position that the designers use engineering judgment in determining if the effect will significantly impact design.
That is, if the stress levels are low, the designer does not transfer the loads. For Cygna's assessment, see Observation PS-05.
I T /16" j 6x6xi TS (TYP) ,
_l._ _ _ _ _ _ _ _ - -
Beam 1 g/
I / 5/16" Weld Detail i _
I I "B"* ^
CG of Weld
{ [ Beam 2 "A"* :
g CG of Beam 1
-t--
i .
- 5. Use of .6Fy for U-Bolts In designing the larger, non-standard U-bolts (i .e., 2-3/4" diameter rods, 2-1/4" diameter rods), the pipe i support organizations have used an allowable tensile. stress of .6Fy; actual bolt stresses were based on 1 the tensile area of the threaded region. Titis conforms with the ASME Code Section III, Appendix XVII, Paragraph XVII-2211. In order to provide further justification for this procedure, ITT Grinnell performed
, a test program for 1/2" diameter and 1" diameter U-bolts (Reference Attachment to TUGC0 letter dated May
- 2, 1984). Based on the results of those tests, ITT has shown quite clearly that .6Fy is an acceptable
- tensile stress allowable for U-bolts.
}
i
, _ 7 - -L ; ,
- Texas Utilities Electric Company; rk=4 bmJk- Independent Assessment Program, Phase 3 lillllilllhl!hilliliffilllll Final Report TR-84042-01, Rev. 1 Page 3 of 9
- 6. Tubesteel Prying on Richmond Inserts In the designs which employ tubesteel/ Richmond insert combinations, Cygna noted that the engineer released the rotation about the Y and Z axes (see sketch) in the STRUDL model for the frame. While release of the
, Y rotation is appropriate since the bolt is free within the tube, release of the Z rotation assumes that the tube will not bear against the washer at point "A" and create a load due to prying on the bolt. TUGC0 has provided justification for this and other analytical assumptions (i.e., the bolt does not carry any .
load in bending; the effect of bolt hole offset on bolt lead) by performing both testing and analysis.
Details of the justification may be found in the TUGC0 letter dated 5/8/84 and in the " Affidavit of John C. Finneran, Jr., Robert C. Iotti and R. Petu Deubler Regarding Design of Richmond Inserts and their .
Application to Support Designs." fc the ietter to Cygns, TUGC0 shows that prying due to rotation about the Z axis is not present when only vertical loads exist. When torsional inments (M x ) exist, the study done by TUGC0 shows that even with small amounts of torsion (1000 in-lb vs 40000 lb tension load), the effect of prying is due to torsion, with no contribution from moments about the Z axis. For large i torsional loads (4000 in-lb vs. 2000 lb tension), the same effect holds true. Cygna then reviewed all tubesteel/ Richmond insert joints within their scope and determined that the configuration analyzed by TUGC0 (4 x 4 x 3/8 TS with 20" bolt spacing) is representative of tne most flexible configurations and, therefore, most conservative. As a result, Cygna finds the method used by TUGC0 to model these connections is acceptable.
i 4Y 1
l U Coeerafa,
/
(q#; /
- r -
t s I = li l
Axiol LW y blL8 h_ *.'l hg'sy), Ma
- "A f mesmrsswa m% .
Milj g'[] Texas Utilities Electric Company; Independent Assessment Program, Phase 3 llll11llllll!!ll11111111llll11 Final Report TR-84042-01, Rev. 1 Page 4 of 9
- 7. Support Self-Lieight Excitation As Cygna found in the Phase 2 review, the design organizations at CPSES do not usually consider additional support load due to the seismic excitation of the support mass in the unrestrained direction. In the case of simple support combinations, such as clamps, struts, and base plates, the effect 4 minimal since the mass is very small. In the case of frames, Cygna has found some examples where self-weight excitation was considered, usually by applying 1.09 in all 3 directions. However, this practice was not commonly employed in the supports which Cygna reviewed. Since the issue of self-weight excitation has been raised and reviewed by the NRC (reference the NRC SIT Report, Item 3h), Cygna did not perform any additional ~
technical evaluations. Cygna did note that the mainsteam supports inside containment involve fairly massive frames, although the applied loads are already sufficiently large such that the added effect may
- be minimal.
- 8. Effect of Support Stiffness As noted in the Phase 2 Cygna Report, the design organizations do not calculate actual support stiffnesses for Class 2 and 3 piping systems. Rather, they limit deflections of frames to 1/16" and do not consider the deformation of standard components, such as struts, clamps and snubbers, or the base plates. Si nce the effects of support stiffness on the piping analysis has been raised by the NRC (the NRC SIT Report, Item 3j), Cygna did not perform a techaical evaluation of this concern other than to note it is necessary to consider the effects.
- 9. Cross-Sectional Properties for Tubesteel In the review of pipe supports, Cygna noted that two of the pipe support design organizations, NPSI and ITT Grinnell, use cross-sectional properties.for tubesteel from the AISC Manual, 7th Edition. Another design organization, PSE, uses the properties from the AISC Manual, 8th Edition. When Cygna questioned the apparent inconsistency, TUGC0 referenced the " Affidavit of J.C. Finneran and R.C. Iotti Regarding CASE's Allegation Involving Section Property Values." As explained in that filing, the tubesteel at CPSES is A500 GRB, cold-formed, for which the section properties from the 8th Edition of the AISC Manual are more appropriate. The differences in section properties between the two editions are minor and have negligible impact on design. As further noted in the TUGC0 response to Cygna (TUGC0 letter 6/8/84), TUGC0 will issue a DCA to specification 2323-MS-46A to note this exception to the AISC 7th Edition. Cygna considers this question adequately addressed and the matter closed.
._ A % =
i=
Q' g k b} Texas Utilities Electric Company; Independent Assessment Program, Phase 3 lllllllllllll1llllll11lll11111 Final Report TR-84042-01, Rev. 1 Page 5 of 9
- 10. " Cinched" U-Bolts on the Component Cooling Water System In reviewing the pipe supports for the Component Cooling Water System, Cygna noted a. number of instances where a U-bolt is tightened around a pipe to provide stability for the support. Cygna asked TUGC0 to provide justification that the U-bolt would not be overstressed. In response to Cygna's request for one
- example, TUGC0 provided calculations in' their 6/8/84 letter and subsequently revised them on June 18, 1984. Cygna has reviewed the TUGC0 calculations and agrees that there will be no adverse stress effects in the U-bolt for the component cooling water systems. For the pipe, see Note 12, which discusses the Westinghouse test and analysis program for U-bolts.
- 11. Rear Bracket Dimensions ,
In reviewing the designs at CPSES, Cygna used the certified vendor catalog and load data available at the site. In Revision 17 of the Design Report Summary (DRS) for rear brackets (ITT Grinnell), Cygna noted dimensions which did not agree with those used by the support designers. The use of larger dimensions would af fect weld lengths and, therefore, design. As explained by TUGC0 in their 6/8/84 letter, Revision 16 of the DRS is the appropriate revision for the dimensions since the majority of the brackets were purchased prior to the issuance of Revision 17 in April,1983 TUGC0 provided Cygna with a copy of
, Revision 16 and Cygna verified that the dimensions used correctly correspond to Revision 16. To further i confirm the appropriate dimensions, Cygna measured rear brackets in those supports chosen for a latter walkdown and confirmed that the installed bracket dimensions are the same as those in the DRS revision used by the designer. Based on the outcome of that walkdown, Cygna considers this matter adequately
- addressed.
- 12. " Cinched" U-Bolts: Effects on Piping, Stability, and the U-Bolt i In reviewing supports on the Main Steam and other systems, Cygna noted instances where a U-bolt was i tightened around the pipe. This was typically done by TUGC0 to provide stability for the support by having the U-bolt act as a clamp. Cygna asked TUGC0 if the local stresses in the pipe and the additional
- stresses in the U-bolt had been considered during the design process. In response to this same question by the ASLB, TUGC0 had contracted Westinghouse to perform a test / analysis program. The details of this program are described in Westinghouse letter EQ&T-EQT-737, dated 3/5/84 i
^
]'y
' l y Texas Independent Utilities Electric Company; Assessment Program, Phase 3 lillllllllllllllllllilllHilli Final Report TR-84042-01, Rev. 1 Page 6 of 9
i The objectives of this test / analysis program were to ensure that:
- 1) ' Stress levels in the U-bolt remained within acceptable limits;
- 2) Stress levels in the piping remained within acceptable limits;
- 3) Stress levels in the crosspieces ~ remained within acceptable limits;
- 4) The U-bolt would maintain the support in a stable configuration (i.e., would'not slip) under maximum allowable strut / snubber angularity (5*);
- 5) The U-bolt would maintain its stability characteristics over time (i.e., would not relax). ,
- 6) The U-bolt would maintain its stability characteristics under normal vibration loading.
As part of the program, TUGC0 selected the following piping: ,
a) 4" sch 160 (stainless) with temperature = 559*F-b) 10" sch 40 (stainless) with temperature = 210"F c) 10" sch 80 (carbon steel) with temperature = 210*F d) 32" with T = 1.25" (carbon steel) with temperature = 557*F These represent a broad range of piping and material combinations at CPSES and would provide assurance
- that the worst combination of wall thickness, pipe size, and temperature effects have been considered.
i At the time of Cygna's review, only preliminary results from this study are available (reference, EBASCO letter dated 6/15/84 from R.C. Iotti to N.H. Williams). Cygna is continuing with an evaluation of this design and will make the results available at a later date. Cygna considers this an open item in this Phase 3 report and finds all supports utilizing " cinched" U-bolts acceptable contingent solely upon the_
acceptability of that test / analysis program. A list of those supports utilizing " cinched" U-bolts for
- stability is provided below
- Support Number Checklist No.
CC-1-020-001-A33K PS-009 i
CC-1-028-007-S33R PS-017*
CC-1-028-701-A33R PS-036 MS-1-001-003-S72R PS-069 MS-1-001-004-S72R PS-070 MS-1-001-005-S72R PS-071 MS-1-002-003-S72R PS-082 i
! Texas Utilities Electric Company; nmun.m " .ggggyyy Independent Assessment Program, Phase 3 Final Report TR-84042-01, Rev. 1 Page 7 of 9
b Support Number Checklist No.
MS-1-002-005-S72R PS-084 MS-1-003-003-572R PS-099 NS-1-003-004-S72R PS-100 MS-1-003-005-S72R PS-101 MS-1-004-003-S72R PS-119
- MS-1-004-005-5/2R PS-121
- Support design revised per TUGC0 letter 6-8-84.
- 13. Embedded Plate Design 4 During the review of supports attached to embedded plates, Cygna noted that in most cases the designers assume a fixed joint at the embedded plate. The governing criteria in Appendix 4 of G&H Specification 2323-MS-46A states that the connections to embedded plates shall be assumed " pin" joints (i.e., forces
. only, no moments) unless' stiffeners are provided, but no guidelines are given for these stiffeners. The standard procedure at CPSES is to assume that the attachment to the plate, usually a beam or base plate, provides the stiffener for the embedded plate. The moments are then distributed to the bolts using a conservative estimate for the dimension of the attachment. Also, in these cases, the lower allowables fer the embedded plates are used. Cygna did find a case in which TUGC0 performed a finite element analysis of the connection to the embedded plate, when their initial approach was too conservative. Based on Cygna's review of the design of connections to embedded plates, we find the approach acceptable.
- 14. A563A Nuts with High Strength Bolts In certain supports at CPSES, Cygna noted the use of A563 grade A nuts with high strength A193 B7 thru bolts. The ASTM specification states that A563 grade A nuts are suitable only for low strength A307
- bolts, based on a comparison of yield and ultimate strength data. TUGC0 has stated that their standard practice is to use A194 2H nuts with A193 B7, but they do allow the use of double A563 grade A nuts, since
- they will have sufficient strength to ensure the acceptability of the joint. Also, all nuts are tightened
" snug tight", thus ensuring both nuts will share the load. In all supports within the Cygna scope, CPSES i designers did use double nuts wherever AS63 grade A nuts were specified for A193 B7 bolts. Thus, the bolted joint design is acceptable.
Texas Utilities Electric Company; L*b l'd [ EA1 Independent Assessment Program, Phase 3 HillHHlHHHillHHlHHH Final Report TR-84042-01, Rev.1 Page 8 of 9 i
l
~
15.1-1/8" Bolt Holes Used in Base Plates with Hilti Kwik-Bolts and in Tubesteel with Richmond Inserts Paragraph NF-4721(a) of the ASME B&PV Code,Section III, provides guidelines for the fabrication and installa-tion of bolting. In it, the Code allows 1/8" oversize bolt holes to be used with 1" bolts made from low strength (yield n 80 ksi) material. Since the tubesteel/ Richmond insert conbinations seen by Cygna use A36 4 threaded rod (yield = 36 ksi), this provision is met. Hilti Kw'ik bolts, however, have a yield greater than 80 ksi, so, in the absence of manufacturer's guidelines, paragraph NF-4721(b)-1 should apply. This paragraph does j not prohibit the use of oversize holes with high strength bolting. As noted by TUGC0 in their 6/8/84 letter, this interpretation was agreed to by both the CPSES constructor and the authorized nuclear inspector. In .
addition, the Hilti Product Management Brochure for Hilti installation states that the wedge clearance hole in a base plate should be 1.17" for 1" bolt, to facilitate installation. Therefore, the use of 1-1/8" holes for Hilti bolts does meet the manufacturer's guidlines. Based on the above, Cygna concurs with the bolt hole -
j diameters used at CPSES.
- 16. Box Frames with 0" Gap l
l In the Phase 3 support review, Cygna noted rare instances where a box frame was used with a strut in place of a pipe clamp. In these cases, the drawing specified a 0" gap between the pipe and frame. Cygna asked TUGC0 to i evaluate the stresses in the pipe and frame, due to thermal expansion of the piping. In response to this and a j similar question from the ASLB, TUGC0 performed calculations on these Component Cooling Water frames; these l calculations shew that additional stresses in the pipe are less than 10 ksi and that additional support loads
! are less than 500 lbs. Since the loads are thermally induced and, therefore, self-limiting, both of these additional effects are well within Code allowables for self-limiting loads. Cygna has reviewed the TUGC0 i calculation (Attachment B to the TUGC0 6/8/84 letter to Cygna) and concurs with the conclusions in that calcula-i tion. In addition to the TUGC0 calculation for box frames or, the CCW system, Cygna performed calculations for a hox frame witn 0" gap located on the Residual Heat Removal / Safety Injection system. This application of the box
~
j frame design was discovered during the Phase 2 pipe support reviews. These calculations concluded that the 16" pipe and the box frame are within code allowables for temperatures as high as 350' F. According to TUGC0 and
.' the NRC Special Inspection Team, box frame designs are primarily found on low temperature systems such as the
- CCW or Service Water system. Discussions with TUGC0 indicate the above application on the RHR system is isolated. No examples of box frames with 0" gap were found on th9 Main Steam system.
4
- 17. U-Bolts Used on Trapeze Supports In a number of trapeze supports reviewed in Phase 3, Cygna noted the use of a U-Bolt to keep the pipe positioned on the frame. In these cases (typically spring supports), there is no upward load on the U-bolt. In effect, j the U-bolt is not needed as a load carrying member, but only to keep the pipe in place on the trapeze beam. In j these cases, Cygna has referenced this note on the checklist to help explain the U-bolt's function.
f 1_ :_*
i [qfej l Q Texas Utilities Electric Company; lllllllll1111111lllll111111111 Independent Assessment Program, Phase 3 i Final Report TR-84042-01, Rev. 1 Page 9 of 9
.