ML20236X830

From kanterella
Revision as of 15:03, 17 February 2021 by StriderTol (talk | contribs) (StriderTol Bot insert)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search
Rev 0 to Issue Specific Action Plan V.B, Improper Shortening of Anchor Bolts in Steam Generator Upper Lateral Supports
ML20236X830
Person / Time
Site: Comanche Peak  Luminant icon.png
Issue date: 10/29/1987
From: Beck J
TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC)
To:
Shared Package
ML20236X826 List:
References
NUDOCS 8712100302
Download: ML20236X830 (63)
v • d • e


Text

{{#Wiki_filter:- _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . i Q) l l l COMANCHE PEAK RESPONSE TEAM RESULTS REPORT l i j l ISAP: V.b

Title:

Improper Shortening of Anchor Bolts in Steam Generator Upper Lateral Supports I REVISION 0 I J l f% 

                                                   /M Issul Coordinator iehi/e9 Datl    /
                                                                  'J         ,

wm ~ ll . ' ' - Io f2 f fi7 Rev M Team'Leade'r .. Date / ' 

                                                                                           /0 f24fS 7 Jot [/ W. Beck, Chairman CPET-SRT             Date             1 8712100302 871207 PDR  ADDCK 05000445
    ,q                                A                  PDR N.A

Revision: 0 Page 1 of 62 RESULTS REPORT ISAP V b Improper Shortening.of Anchor Bolts in Steam Generator Upper Lateral Supports

1.0 DESCRIPTION

OF ISSUE' IDENTIFIED BY NRC Technical issues and certain allegations with potential safety implications concerning improper shortening of anchor bolts in steam generator upper lateral restraints (SGULs)* were communicated by the NRC to TU Electric. The following assessment of this issue was described by the TRT in SSER 10, Page N-328'(Reference 9.1). The TRT was informed that sosae anchor bolts in the steam generator upper support beams were shortened during installation to less than the length shown on the design drawing without proper authorization. The TRT was told that the bolt cutting incident occurred either because the hole of the anchor device was filled with debris, or the threaded portion of the bolt had concrete mix stuck to it. There are 18 bolts at each end of each of four beams, totalling 144 bolts. There is one beam for each steam generator. The bolt 

    /

threads into an anchor device embedded in the concrete wall. [(]_/ The acceptable bolt length or the length of bolt available for threading into the anchor device is vital to ensure structural capability of the support beams. The TRT attempted to review TUEC reco:cds for Ultrasonic. (UT) measurement results and general installation practices. The TRT was told that Ultrasonic-testing of these types of bcits was not a procedural requirement; however, TUEC was unable to provide any other installation records for TRT review. The TRT concludes that such unauthorized bolt cutting and lack of installation inspection records is a violation of QA procedures and Criterion XVII in Appendix B of 10CFR50. Since the support beams are essential to provide lateral restraint for the steam generator during a LOCA or seismic event. l adequate anchoring capability of the bolts has safety significance and, as a result, appropriate measures are needed to ensure confermance with General Design Criterion 1 of

l. 10CFR50.

l l

  • The word support is eometimes used instead of the word restraint. As ,

the SGULs do not provide support for the dead load of the steam j generators but only restrain the motion of the steam generators under O s seismic or LOCA loads., the use of the word rostraint is more appropriate.

M Revision: 0 Page 2 of 62 RESULTS REPORT ISAP V.b (Cont'd) 2.0 ACTION IDENTIFIED BY NRC-The TRT (Reference 9.1, Ft;ges N-328 and N-329) identified that the following Svtion should be taken on this issue: Accordingly, TUEC shall provide evidence, such as Ultrasonic measurement results, to verify acceptable bolt. length. Should unauthorized bolt cutting be verified, TUEC shall: Replace shorrened bolts wit.h bolts of proper length, or provide analysis to justify the adequacy of shortened bolts as installed. Provide justification or-propose measures to ensure that no s!.milar concern exists for bolting.

3.0 BACKGROUND

I There are four SGULs in each Unit; one in each steam generator b compartment (see Figure 1)*. A SGUL consists of a steel box beam, with end plates welded to its ends; two embedments that are poured into the steam generator compartment concrete structure; and a shim pack at each SGUL beam-to-embedment connection (see Figura 2). The SGUL beam is connected at both of its ends 'to an embedme it with - eighteen bolts that go through the holes in the beam end plates and the shim pack into the threaded holes in the embedment (see Figure 3). The bolts have a nominal diameter of 2-1/2 inches, are made of A-540 cieel, and were tightened to a snug tight condition to provide a bearing connection of the SGUL beam to the embedment. 1 Table 17A-1 of the FSAR specifies the SGULs as Seismic Category I, I safety-related components. The TRT investigated the allegation of unauthot1 zed anchor bolt l shortening for the SGULs. As part of that investigation TU Electric was requested to provide the installation and Inspection Records for the subject SGULs; however, the records could not be located in the permanent plant records vault. Nonconformance Report, NCR-M-84-100384 vas icsued by TU Electric on September 6, 1984 to document the lack of Inspecticn Records. l l

  • There is also a steam generator lower lateral restraint in each steam .

generator compartment. The lower lateral restraints were reviewed l within DSAP VIII as part of the Design Adcouacy Program.

Revision: O Page 3 of 62 RESULTS REPORT ISAP V.b (Cont'd)

3.0 BACKGROUND

(Cont'd) The disposition of the Nonconformance Report required inspection of the Unit 1 installation. This inspection included UT measurement of bolt lengths to determine if the design bolt thread engagement into the embedded plate was provided. The inspection was performed in accordance with Operation Travelers CE-84-192-8902, CE-84-193-8902, CE-84-194-8902, and CE-84-195-8902. Results from the measurements indicated that some bolts had inadequate thread l engagement. This Discrepancy was reported on January 17, 1985 to the NRC in accordance with 10CFR50.55(e). 4.0 CPRI ACTION PLAN The objectives of this action plan were to ensure that the SGUL bolts meet design requirements, to assess causes that led to the improper installation of SGUL bolts and lack of inspection documentation., and to review similar installation circumstances to determine if the issue pertains to other bolting installations. The steps to achieve these objectives are discussed in the O: following subsections. 4.1 Scope and Methodology 4.1.1 Inspections and Design Revisions of SGUL Connections Based on the inspections perfolted on Unit I for some of the bolts, thread engagement was determined to be less than required, (as identified in Section 3.0). Gibbs & Hill (G&H) reevaluated the design and loading conditions. This led to reanalysis of the steam generator compartment response by G6H and of the reactor coolant system response by Westinghouse. The reevaluation of the design provided the basis for reinspection and rework of the Unit 1 SGUL bolts. A similar inspection of Unit 2 has been initiated. The torrective actions taken for Unit I will also be perfsrmed on Unit 2 as determined applicable based on the inspections. 4.1.2 Review of Fabrication and Installation of SGULE TFa bolts and the blind threaded holes of the embedmer.t were examined during the process of bolt replacement to assist in the assessment of root cause. Those aspects of the fabrication and installation process that might have directly or indirectly influenced the as-built w _ _ _ _ _ - - - . - - . _ . - - _ _ - - - _ _ _ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ~ - - - - - - - - _ - ' ^ ~ '

Revision: 0 Page 4 of 62 

 ^

RESULTS REPORT l O ISAP V.b h (Cont'd) I' 4.0 CPRT ACTION PLAN (Cont'd) thread engagement were also reviewed. This includes the review of dimensional tolerances of parts in the assembly, such as plate thickness, shim sizes, concrete placement, and bolt length. 4.1.3 Inspection of Other Populations of Bolted Connections To assess the potential generic applicability of the SGUL bolt thread engagement issue to bolted connections dependent on thread engagement, other installations were inspected on a sampling basis. Two populations with the feature of bolts being installed in blind holes, as in the SGUL connection, were inspected. 4.1.3.1 Inspection of Richmond Inserts Richmond inserts were used extensively in pipe support applications. The p populations of Richmond inserts that were to 5 be sampled for inspection of thread engagement were those utilized'in ASME pipe supports in Units 1 and 2. The sampling plan was designed in accordance with Appendix D of the CPRT Program Plan. The objective was to either provide reasonable assurance that the  ; inserts had design engagement or if l discrepancies were detected, they were'to be investigated and dispositioned. It was determined at the inception of this task that a similar, but more encompassing, investigation of thread engagement in Richmond inserts was being conducted as part of the QA/QC Review Team ISAP VII.c. 

                                                             " Construction Reinspection / Documentation Review Plan". It was decided that this investigation satisfied the ebjective of the investigation planned under this ISAP and the results were adopted for use in satisfying the purposes of this ISAP.

4.1.3.2 Inspection of Drilled and Tapped Blind Hole Bolteo Connections f Based on review of bolted connections where thread engagement is an important design attribute, the population of structural L_______-. __

Revision: 0 Page 5 of 62 RESULTS REPORT j ISAP V b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd) bolted connections with a drilled and tapped blind hole

  • was selected for inspection. A sampling plan' developed in accordance with the Appendix D of the CPRT Program Plan was utilized. The thread engagement for the sample was determined in accordance with an inspection plan developed for this type of bolted connections. Results of this inspection were reviewed for compliance with the design.

4.1.4 Review of Procedures for Documentation, Inspection Criteria, and Quality Control A third-party review was performed of the process by which critical attributes of bolted connections were specified and their associated inspection criteria and documentation requirements defined. The initial scope. 

        ')                                                                                   of this effort was focused on an evaluation of the s.,/                                                                                   circumstances'that led to the lack of appropriate records for the SGULs. This activity was to be ex} anded to other areas if it was- determined that identified program deficiencies may be applicable elsewhere or if problems were identified as a result of the inspections noted under Section 4.1.3. The requirements for engineering approvals and field controls for bolt c9tting were also reviewed by third party.

4.1.5 Investigation of Thread Damage While removing the SGUL bolts in Unit I during the inspections referred to in Section 3.0, thread damage was observed in several embedment holes. Subsequently, in response to NRC Staff's recommendation, a task was added to the scope of the ISAP V.b to assess the possible generic applicability of the thread damage problem. After review of the issue, the third party performed an fnvestigction of causes for the thread damage in the SGUL embedment holes and an evaluation L Hereafter referred to as drill and tap blind connections. 1 l

I Revision: 'O i Page 6 of 62 _RESULTS REPORT l' v ISAP V.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd) 1 of whether these causes were applicable to the population of structural drill and tap blind { j connections. Further, threaded rods were removed from  ! Richmond inserts used in voided pipe' supports and ) inspected for thread condition. , A third-party overview of all aspects of the action plan was conducted. The inspections of the SGUL bolts and threaded holes were overviewed. Design calculations that were performed to evaluate the SGUL bolt engagement requirernents were reviewed. The inspection plans for all inspections of Richmond inserts and the drill and tap blind connections were reviewed and the performance of these inspections was 4 overviewed by the third party. 4.2 Participants Roles and Responsibilities The organizations and personne1'that participated in this effort. are listed below with their respective scopes of work. 4.2.1 TUGC0 Nuclear Engineering (TNE) Civil / Structural Discipline 4.2.1.1 Scope Prescribed SGUL connection initial inspections Reviewed SGUL connection initial inspection results Selected samples of Richmond inserts and drill and tap blind connections 4.2.1.2 Persom-si Mr. C... Hooton TNE Civil / Structural  ; Discipline Supervisor i- Mr. T. L. Wright Civil Engineer n f Mr. C. L. Corbin Civil Engineer 

                                                                                                                                              )

[O

v-Revision: 0 Page 7 of 62 RESULTS REPORT ISAP V.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd) 4.2.2 Gibbs & Hill, Inc. 4.2.2.1 Scope Prepared calculations to evaluate adequacy of the as-built SGUL bolt thread engagements Performed reanalysis of the steam generator compartment response to seismic and LOCA loads 4.2.2.2 Personnel Mr. E. B. Bezkor Structural Job Engineer Mr. A. M. Kenkre Structural Squad Leader 4.2.3 Brown & Root, Inc. 4.2.3.1 Scope Implemented bolt replacement rework Removed voided pipe supports which utilized Richmond inserts in preparation for inspection of thread condition of threaded rods used in the inserts 4.2.4 TUGC0 QC 4.2.4.1 Scope Performed measttrements to determine SGUL bolt lengths Participated in inspections of SGUL embedtent bolt holes directed by Westinghouse O

Revision: 0 Page 8 of 62 

    .\,s!(}                                  RESULTS REPORT ISAP V b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd)

Performed measurements utilizing UT and other techniques to determine the thread engagement lengths for the sampling inspections of Richmond inserts and drill and tap blind connections Inspected the thread condition of threaded rods removed from voided pipe supports 4.2.4.2 Personnel Mr. C. T. Brandt Technical Assistant to the Director. Quality Assurance 

        ,-s,                4.2.5 Westinghouse Electric Corporation i        \

k/ 4.2.5.1 Scope Inspected and evaluated SGUL embedment hole thread condition i Performed reanalyses of reactor l coolant system response to seismic -{ and LOCA loads Performed reanalysis of SGUL connection Effected design revisions for SGUL connection I Effected SGUL embedment hole thread l repair 4.2.5.2 Personnel l Mr. R. E. Kelly Manager, Structural l Engineering l l Mr. H. L. Ott Senior Engineer l f/~Sc. l N sl l

i Revision: 0 Page 9 of 62 l 

        /~ T                                                                  RESULTS REPORT
       'w)

ISAP V,b (Cont'd) I 4.0 CPRT ACTION PLAN (Cont'd) 4.2.6 Stone & Webster Engineering Corporation (SWEC) SWEC was assigned on October 24, 1986 as the lead contractor with the responsibility to perform the Project engineering activities associated with this ISAP. 4.2.6.1 Scope Reviewed and finalized the sample selection of Richmond inserts and drill and tap blind connections for the thread engagement inspections Prepared inspection plans for the measurements to determine the thread engagement length of the samples of 

         ,g                                                                         Richmond inserts and drill and tap
        'q       )                                                                    blind connections Prepared inspection plans for the inspections of thread condition of threaded rods used in voided pipe supports Evaluated inspecticn results Reviewed the installation process        '

and documentation for 15 subpopulations of drill and tap blind connections to arsess the possibility of thread damage. 4.2.6.2 Personnel Mr. R. D. Ciatto Assistant Lead Structural Engineer Mr. T. W. Houston Supervising Structural Engineer Mr. M. P. Holland Group Supervisor, 1 

           ,-s                                                                                     Structural Division I        i t
         \m/                                                   .

T l Revision: O Page 10 of 62 

   '~ 's (v     4 RESULTS REPORT ISAP V.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd) 4.2.7  Third Party 4.2.7.1    Scope Reviewed inspection plans for SGUL                    l bolts Reviewed the original design calculations for the SCUL connection Reviewed process of engineering specification of, and procedural and QC inspection requirements for, bolting attributes Reviewed sample selection of Richmond inserts and drill and tap
  ,~s                                         blind connections t
  \- -)                                 -

Reviewed inspection plans for Richmond inserts and drill and tap samples Reviewed the process of development of final design calculations i Analyzed root cause and generic l implications I Prepared results report Overviewed inspections of SGUL bolts and embedment holes (SwRI) Overviewed inspections of Richmond inserts and drill and tap blind connections (SwRI) 4.2.7.2 Personnel Mr. H. A. Levin TENERA - CPRT Mechanical Review Team Leader d l l

f 4 Revision: 0 Page 11 of 62 f[ U RESULTS REPORT V ISAP V.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd) Dr. J. R. Honekamp TENERA - CPRT TRT Issues Technical Manager Mr. J. C. Miller TENERA - CPRT TRT Issues Manager Prof. M. J. Holley TENERA Consultant, Hansen, %s11ey & Biggs Dr. W. Munsa TENERA Consultant - University of Illinois Dr. J. Arros TENERA - Issue Coordinator 

     ,~

4 ' 'j Mr. G. L. Lagleder Southwest Pesearch Institute (SwRI) Dr. F. A. Webster Jack Benjamin & Associates (JBA) - CPRT Statistical Advisor 4.3 Personnel Qualification Requirement Where tests or inspections require the use of certified inspectors, qualifications at the appropriate level were to the requirements of ANSI N45.2.6, " Qualification of Inspection, Examination, and Testing Personnel at Nuclear Power Plants." CPRT third-party inspectors were certified to the requirements of the third-party employer's Quality Assurance Program, and trained to the applicable inspection procedures. Third-party participants in the implementation of this Action Plan met the personnel qualification and objectivity requirements of the CPRT Program Plan and its implementing procedures.  ! Other participants were qualified to the requirements of the l,h CPSES Quality Assurance Program or to the specific 

      \~ /                   requirements of the CPRT Program Plan. Activities performed by other than third-party personnel were governed by the applicable principles of Section III.K, " Assurance of CPRT Program Quality", of the CPRT Program Plan.

[ _ _ _ _ - - _ _ _ _ _ _ _ _ _

Revision: 0 Page 12 of 62 

 *yb                                                                RESULTS REPORT L\.j ISAP V.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd) 4.4   Procedures The initial inspections of the SGUL bolts and embedments were performed in accordance with the Project operation travelers identified in Section 3.0. The UT measurements of the bolt lengths performed as part of these inspections were performed in accordance with the Project QC Instruction QI-QP-11.18-3,
                                              " Ultrasonic Measurement Thickness", Revision 4 (Reference 9.2).

The inspections of tha SGUL embedments directed by Westinghouse were performed in accordance with Project 

                                            -operation travelers (Reference 9.3).      Installation of new shims and bolts was also performed in accordante with Project operation travelers (Reference 9.4).

The " Procedure for Evaluation of Thrcad Engagement Length Of Other Blind-Hold Installations" contained in Appendix G, and n{'j- 

                                             " Inspection Service Request Fozu Procedure for Other Blind-Hole Installations" contained in Appendix H of Significant Technical Issue Report (STIR) STIR-CPRT-S-001 (Reference 9.5) provide the plan for, and results of, the evaluation of the thread engagement adequacy of the population of drill and tap blind connections (see Section 4.1,3.2). The sampling plan for this task, which ie in accordance with the Appendix D of the CPRT Program Plan, is contained in Appendix B of the STIR. The UT measurements of the bolt lengths were performed to the Project QC Instruction QI-QP-il.18-3. The inspection of thread condition of threaded rods inserted in the Richmond inserts used in voided pipe supports (see Section 4.1.5) was performed in accordance with the plan contained in Appendix L of the STIR.

The third party reviewed and approved all travelers and procedures utilized in the inspections to gather information for the purposes of this ISAP, (Reference 9.6). , 4.5 Standards and Acceptance Criteria j Conformance of actual SGUL bolt thread engagement lengths to l 1 the required engagement lengths as specified in the design I determined the acceptability of the connection installation. { O The acceptance criteria for the inspections of the thread ( C/ engagement length of bolts or threaded rods installed in the l Richmond inserts was the requirement of two times the diameter of the bolt, or rod, plus 1/8 inch as specified in the Project l Specification 2323-SS-30, " Structural Embedments".  ! e_-- - - .

Revision: 0 Page 13 of 62 h [G RESULTS REPORT ISAP V.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd) The acceptance criterion for the inspection of the thread engagement length of bolts in the drill and tap blind connections was the thread engagement required on design drawings. For those cases where the thread engagement was not specified on the design drawings, the acceptance criterion was: a thread engagement equal to the nominal bolt diameter, which was established by SWEC in accordance with industry practice, as a conservative design requirement for these es, es. 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS The implementation. of the ISAP V.b consisted of inspection and review of the design of the SGUL connections (Section 5.1); review of fabrication and installation of the SGUL assemblies (Section 5.2); inspection of thread engagement in two populations of bolted connections (Section 5.3); review of the process of engineering 

   '                                                                     specification of, and procedural and QC inspection requf.rer.ents

' for, bolting attributes (Section 5.4). Based on the observation of thread damage in the early inspections of the SGUL connections, an investigation of the thread damage and its possible generic applicability to other bolted connections was added to the scope of this ISAP (Section 5.5). A summary of DIRs is provided in Section 5.6; a safety significance evaluation of devistions in Section 5.7; and a discussion of root cause and generic implications in Section 5.8. 5.1 Reinspection and Redesign of SGUL Connections Both the inspections and design analyses of the SCUL connections were performed in two phases. The first phase of activities, which included both inspections and design evaluations, was conducted from January 1985 through March 1986, and was directed by TNE. These activities focused on determination of both thread engagement requirements and

actual thread engagement of the originally installed bolts. l

' Thread damage in some of the SGUL embedment holes was observed during these inspections (DIR D-2476). As a part of these ' activities G6H performed analyses to evaluate the adequacy of the SGUL connections with the conclusion that the originally installed bolts were not adequate. Whereas design calculations had been performed to size the SGUL beams, no () design calculations for the SGUL bolts that had been generated prior to 1985 could be located (DIR D-2484).

Revision: 0 Page 14 of 62 () RESULTS REPORT ISAP V b (Cont'd) l 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) The second phase took place in early 1986, when Westinghouse was assigned the responsibility of resolving the problems in the SCUL connections. This task was not part of Westinghouse's responsibility as the NSSS supplier, but a separate task. In April 1986, the second set of inspections to evaluate the thread damage in the embedment holes were conducted under Westinghouse direction. Based on this evaluation. Westinghouse revised the design of the SGUL' connections in December 1986. 5.1.1 Inspection of SGUL Connections Inspections Directed by TNE In response to the TRT requests, the first task implemented was inspection of the SGUL bolts. The inspections were performed by Project QC in accordance O with operational travelers prepared by the Project (see Section 3.0 for traveler numbers). During the inspections conducted from February 1985 through March 1986, all of the 144 bolts of the four SGUL beame (eight connections) of Unit I were removed. These inspections included determination of insertion length utilizing UT, diameter of the beam end plate holes, and visual examination of the holes in the embedments. The insertion length of the 144 bolts varied from 0.73 inch to 2.39 inches, with an average of 2.02 inches (Reference 9.7). The term insertion length is used here because in some cases the effective thread engagement had been reduced to less than the insertion length due to grinding down of the threads at the end of the bolts (DIR D-2476). While removing the bolts, severe binding of a bolt 4 against an edge of its hole in the beam end plate i occurred in many cases. In some cases, binding of a bolt against an edge of its hole in the shim pack al i occurred. This indicated a problem in the alignment of the holes in the embedment, in the shim pack, and in the beam end plate. Alignment problems were observed ) for most of the bolts (Reference 9.8). In many cases 1 the binding of the bolt was so severe that major O efforts by the craft were required to remove the bolt, i

j i Revision: 0 Page 15 of 62 1 RESULTS REPORT ISAP V.b { (Cont'd) 1 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) Thread damage was observed' in many of the embed holes (DIR D-2476). Based on the type and location of the' , damage within the holes as well as on information on the SGUL insts11ation process (see Section 5.2 below), it.was determined that some of the thread damage in the I holes near the surface of the embedment was apparently due to inadvertent grinding of the embedment holes j while grinding the holes in the shim pack and in the  ! beam end plate to facilitate insertion of the bolts during installation. In many cases there was a localized area (a patch) deeper in the hole where 

                                                                                                                       )

threads had been damaged. Judging from the effort - required to remove the bolts, this damage was most i likely caused when bolts were forced into the misaligned holes and the lead end of the bolt damaged threads in a localized area. In some holes, cross-threading was observed. It was judged by the third party that the misalignment problem was the cause O of this patch-type thread damage and cross-threading (see Section 5.5). In some cases the marks at the bottom of the bolt hole indicated that the bolt had bottomed out. This was due to cutting of the bolt in the field to a length longer than was appropriate. The vendor drawings (Reference 9.9), which had been approved by G6H, indicated that the bolts were to be field cut "as required to fit end plate, shims, and embedment plate". l The length of the threaded depth of the embedment hole was specified to be "2-1/4 inches, minimum (typical)'." No thread damage that could be attributed to the misalignment problem was found on the bolts. This can be explained by the fact that the bolts are made of harder steel than the embedments. However, many bolts ' had been ground at the end, ir some cases over a length of one inch, so that the crests of the threads were flattened. The grinding may have been perfcrmed to facilitate insertion of the bolts into the misaligned holes. It was also observed that after the binding was released, some of the bolts had a very loose radial fit ( when inserted in the hole, indicating that the hole may } ' have been oversized. The Project generated NCRs for all cases where thread damage was observed, where there was indication that the bolt had bottomed out or where O- the embedment hole was oversized. l l

Revision: 0 Page 16 of 62 1 RESULTS REPORT b,A ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) l It was observed by the third party that the pattern of j the locations of binding within the holes of a connection did not suggest that the binding (and misalignment) was due to a simple translation or j rotation of the mating components of a connection. i Rather it suggested that even if an adjustment had been i made in an attempt to match the holes (assuming the connection at the other end of the beam would allow it), such a match could not be achieved. This raised the question of whether the alignment problem was due to a mismatch of mating components that were intended to be assembled with a uniquely specified match of the mating components (with the orientation around the length axis of the beam also specified). This issue is addressed further in Section 5.5.1. The third party overview of these inspections for p Unit I was provided by SwRI and is documented in SwRI surveillance and summary reports (Reference 9.8). Inspections Directed by Westinghouse In early 1986, Westinghouse was assigned the responsibility of resolving the problems in the SGUL connections. This involved evaluation of the thread damage by new inspections performed under Westinghouse direction, analyses of the connections that accounted for the reduction in the bolt pull-out capacities due to thread damage (see Section 5.1.2), and installation of new bolts. These inspections were performed in 1986 in accordance with and documented on Project Operation Travelers (Reference 9.3). The inspection was performed by Westinghouse representatives and Project QC, and included measurement of the minor and major diameters of the embedment threads, the length of the threaded portion of the hole, and total depth of the hole. Further, both the angle of bolt inclination and the size of the gap between a bolt of nominal diameter of 2-1/2 inches and the end plate hole were determined. The existence of any thread damage was determined by visual inspection; the location, extent, and type of damage was documented. O

l Revision: 0 Page 17 of 62 O RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) Westinghouse decided, based on their evaluation of the thread damage, to prescribe repair of two holes that had sufficient thread damage to require some_correc;1ve action, given their respective locations, by drilling and retapping them to a 2-3/4 inch diameter. Bolts-with 2-3/4 inch diameter providing 2-1/4 inch thread engagement were specified to be installed in these holes. The embedment holes with minor thread damage-were not repaired; Westinghouse accounted for such damage in the design calculations (see Section 5.1.2) by reducing the thread capacity in these holes. The inspections and repairs were overviewed by the third party.(Reference 9.8). Initial inspections of the Unit 2 SGUls have been performed by the Project, with the third party conducting an overview. The observations are similar to those in Unit 1; i.e., misalignment, insufficient I \ thread engagement, grinding of bolt threads, some dirt / debris in embedment holes, and damaged threads in some of the embedment holes, (Reference 9.10). Accordingly, the corrective measures implemented for Unit I will also be implemented for Unit 2. 1 5.1.2 Design of SGUL Connections The design of the SGUL bolted connections was reviewed as part of the scope of this ISAP. The design of the SGUL beam, the design of the embedments and the analyses of the Steam Generator compartments were reviewed as part of the CPRT Design Adequacy Program within DSAP VIII. A number of discrepancies were identified in those reviews. Appropriate DIRs were written and are documented in Issue Resolution Report DAP-E-C/S-502, " Concrete Design-Reactor Building Internal Structures". The following sections provide a discussion of the design of the SGUL bolted connections. Design Concept of SGUL Connections Two mechanisms may cause loading to the SGUL connections: the impact of the steam generator onto the O SGUL beam due to seismic or LOCA loads and the relative movement of the concrete structure at the two ends of the beam with respect to each other. The seismic and

___ _ _._.______ _ . ~ - _ _ - - - . - - - _ - - - -- - - -- - - - - - - - - - - - - - - - - - - - - - fi Revision: 0-_ _; Page 18 of 62 { l 

                                                                                                                                                                                  )

RESULTS REPORT ISAP V,b l (Cont'd) i 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) LOCA loads cause bending moments and shear and axial forces at the SGUL connection that may impose significant stresses on the bolts. 1 I According to G&H, the initial design concept for the  ! SGULs was to provide essentially pinned end cordstions for the SCUL beam, i.e., the SCUL connections,were to transmit shear loads but not bending moments. The - design calculations for both the SCUL beams and the ' embedment anchorages reflect this assumption. However, the connection detail consisting of 18 bolts around the perimeter of the SGUL beam end plate, tightened to a snug tight condition, does not provide a pinned condition, but a moment-resisting condition at' the connections (DIR D-0003). (Design of the SGUL beam with the pinned-connection assumption led to a conservative design of the beam. Consistent with the pinned-connection assumption, the design calculation i for the embedment anchorage did not consider bending moments at the anchorage. This was incorrect in light of the true behavior of the connection and led to a non-conservative evaluation of the anchorage. 1 Original Design of SGUL Connections No calculations evaluating the structural adequacy of ' the SGUL bolted connections prior to the CPRT investigations could be located, although design calculations had been generated to size the SGUL beam (DIR D-2484) As the SGULs are Seismic Category I, safety-related components, their design should have included evaluation of the connections by appropriate calculations. Generation of such calculations was a G&H responsibility. G6H prepared such a calculation (Reference 9.11) in January 1985. This calculation determined the thread engagement requirement at the location of each bolt in the array of 18 bolts of each l connection. The third party reviewed this calculation i and found it acceptable (Reference 9.12). Comparison of the calculated thread engagement requirements to the thread engagements of the installed bolts, as measured by UT prior to removal of the bolts ' .O l for inspection, revealed that thread engagement was O 1

Revision: 0 Page 19 of 62 RESULTS REPORT ISAP V,b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) inadequate for 14 of the 144 bolts (Reference 9.13). Based on this conclusion, a plan was prepared by the Proj ec t to correct the problem by installing bolts that met the criteria established by the G&H calculation. l (These criteria permitted different engagements at various locations on the connections.) However, due to the subsequent findings of the misalignment problems during the inspections, and discrepancies identified by the third party in the LOCA loads used in the finite element analyses of the steam generator compartments that provided the load information for the evaluation of the SGUL connections, the approach was modified as discussed below. Revision to SCUL Loads The loads to the SGUL connections that were used in the evaluation of these connections were obtained from O finite element analyses of the steam generator compartments that were performed by G&H in 1984. These analyses incorporated all loads and load combinations applicable to the steam generator compartment. The impact of a steam generator onto a SGUL due to seismic or LOCA loads, the effects of LOCA loads on the walls, i.e., temperature and pressure effects, and the effects of temperature on the beam subject the SCUL (and its connections) to significant loads. The following provides a brief description of the analyses of the steam generator compartment response and the reactor coolant system response to seismic and LOCA loads. These analyses provided the load information for the evaluation of the SGUL connections. To perform a seismic analysis of the steam generator compartments and to determine the SGUL loads, G6H first analyzed the building response to seismic loadings. Westinghouse used this steam generator compartment response as an input to their seismic analysis of the reactor coolant system. This analysis, in turn, determined the loads that the steam generator imparts to the SCUL at the point of the steam generator impact. Analogously, for a LOCA analysis, G&H provided the O pressure loads to which the reactor coolant system is subj ected, and Westinghouse used this as input to its

I i Revision: 0 Page 20 of 62 ( RESULTS REPORT ISAP V.b l (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) dynamic analysis of the reactor coolant system, which

  • in turn gave the loads to which the SGUL is subjected at the point of the steam generator impact. In its analysis of the reactor coolant system, Westinghouse incorporated the stiffness properties of various <

support points of the coolant system including the SGUL, as generated by G6H. f 1 ( During third-party reviews of the SGUL issue in early f 1985, discrepancies concerning the LOCA loads were identified in these analyses (DIR D-0003). The G6H finite element analysis of the steam generator j compartments, performed in 1984, provided the load 1 information for the evaluation of the SGUL connections, I but did not incorporate a revision to the LOCA loads issued by Westinghouse earlier in 1984. After review of the LOCA loads, and discussions between g Westinghouse, G&H, and the third party 

     )                   (Reference 9.14), it was further discovered that the LOCA scenario that had been selected for the analysis as the governing scenario did not result in the worst case loads to the SGUL and its connections. Instead, another LOCA scenario (different location of postulated primary loop break) produced higher loads.

l Discrepancies were also identified in G&H calculations that evaluated the structural adequacy of the anchorage of the SGUL embedments to the concrete structures of the steam generator compartments (DIR D-0003). These findings gave rise to a process that involved: 1) significant reanalysis by G&H of the steam generator compartment response and by Westinghouse of the reactor i coolant system response, when subjected to seismic or LOCA loads, and 2) inclusion of the review of the structural analyses related to the SCUL issue, e.g., SGUL beam design, steam generator compartment concrete design and design of the steam generator lower lateral restraint, into the Design Adequacy Program's DSAP VIII l scope. A description of the evolution of the SGUL loads and of the underlying analyses is provided in Reference 9.15. As a result, G6H provided their revised analyses in October 1986 (References 9.16). These analyses provided, among other things, the loads [m used in the Westinghouse reanalysis of the SGUL D} connections. j (

j Revision: O Page 21 of 62-Q RESULTS REPORT O' ISAP V,b (Cont'd) ) 

                                                                                                                     \

I 5.0 IMPLEME!!TATION OF' ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) l Redesign of SGUL Connections Based on the loads to the SGUL as obtained from the revised G&H analyses of the steam generator compartment response and on its inspection and evaluation of the SGUL embedments (see Section 5.1.1), Westinghouse provided a redesign of the SGUL connections in the report " Comanche Peak, Inspection and Evaluation of Steam Generator Upper Support Beam Anchor Bolts", i Revision 2, (Reference 9.17), in December 1986. The redesign involved use of bolts providing a minimum thread engagement of 2-1/4 inches, and, in order to eliminate bending stresses in the bolts, pretensioning of the bolts so that shear forces in the connection are resisted by friction, i.e., the connection was made a friction connection. The third party reviewed the Westinghouse report and found it to be acceptable (Reference 9.18). A further revision to the design reduced the thread engagement requirement to 2-1/8 inches (Reference i 9.19). Third party will review this design change. 5.1.3 Reassembly of the SGULs The reassembly of the Unit 1 SGULs involved I installation of new bolts with washers. The bolts were to be torqued to a pretension, in accordance with the Westinghouse redesign, to provide for a friction connection. While installing new SGUL bolts, gaps in the SGUL connections were discovered between the beam endplates and the shim packs and between the shim packs and the embedments. Because of the gaps, the friction connections called for in the design could not be achieved by the specified bolt pretensioning. The Project decided to replace the shim packs with new shims that are machined to provide a fit with gaps not exceeding 0.005 inches in the immediate vicinity of the bolt holes (References 9.19 and 9.20). The new shims were installed in accordance with operation travelers (Reference 9.4). The third party overviewed the installation.

Revision: 0 Pahe 22 of 62 () 'RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) 5.2 Review of Fabrication and Installation of SCULs To investigate the factors that led to the discrepant-installation of the SGULs and the lack of adequate QC inspection records of the installation, the fabrication and installation history of the SGULs was reviewed (Reference  ; 

                                 -9.21).                                                                                                                                                               -'

Fabrication of the SGULs for Unit I was subcontracted to AFC0 by Brown & Root in 1977 (Reference 9.22). The design requirements were specified by C&H and the design was i established on drawings prepared by G&H. G&H provided these drawings to AFC0 along with the requirement to fabricate the SGULs in accordance with Specification 2323-SS-17, 

                                  " Miscellaneous Steel". AFC0 fabricated the SGUL beams,                                                                                                                '

embedments, and shim packs (see Figure 2). AFC0 also prepared fabrication drawings that were reviewed and approved by G&H.

  • These drawings indicate that each piece in the assembly was
   \                              given a unique " piece mark". During fabrication, each end plate was match-drilled with a particular embedment.

According to the drawings, the end plates and corresponding # embedments were match-marked (different from piece marking) to ensure proper mating. The holes were located using a template and, after drilling, were checked to assure the proper location using a tolerance of 1/16 inch. The shim packs were I drilled separately. These holes were 1/8 inch larger than the corresponding holes in the end plates (Reference 9.23 and 9.24). The bolts used for the end plate-to-embedment connection were supplied by AFC0 as well. When fabrication was completed, the components (including the shim packs) were assembled in AFCO's shop as part of an AFC0 QC inspection. The assembly consisted of inserting and tightening the four corner bolts, and then inserting and hand-tightening the remaining bolts to assure proper alignment (Reference 9.24). This QC inspection was documented on AFC0 

                                 " Fabricated Item Traveler" forms that were included in the shipment (these documents are attached to the Brown & Root Receiving Inspection Report packages, associated with the SGULs). These forms contain checks for items such as "fitup" and " threads", and indicate all items to be satisfactory.

The stamped piece marks indicated on the drawings were O. identified on the support beams during walkdowns performed by the third party (Reference 9.25). Markings on the other components (shim packs and embedments) were not detected during the third-party walkdown. The markings are probably on the surfaces that are embedded in concrete. l l l

Revision: 0 Page 23 of 62 RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF.RESULTS (Cont'd) It is not clear whether the SGUL assemblies were shipped as partially assembled units or as separate components. For only one shipment, the records indicate that a SCUL beam and its matching embedment were included in the same shipment (References 9.26, 9.27, and 9.28). It appears likely that components were shipped unassembled. Evidence in the CPSES Receiving Inspection Reports indicates  ; that the AFC0 drawings were used to identify matched i components. Two inspection reports (part of References 9.26 and 9.27) stated that for two beam-to-embedment matings (identified by AFC0 markings on the Inspection Reports) the 

                " orientation of sleeves to laterals" was "per applicable specifications". Although the AFC0 drawings are not cited, these documents must have been available for the inspector to have made such a determination because the correct matings were shown only on these drawings. It is not clear how the inspector made this determination without the components being assembled.

At the time of receipt inspection onsite, NCR M-906 was-written concerning acceptability of the SCULs. The QC inspector thought that ASME Code applied to the SGULs and that they would require an NF designation. 11y vendors holding an NPT stamp can fabricate such components. Since AFC0 did not hold such a stamp, the components had a QC hold placed on them. The NCR was resolved when it was determined that the SGUls could be classified as non-NF (Reference 9.29). Subsequently, the SGULs were released for construction. There is no documentation to indicate that construction personnel were apprised of the AFC0 drawings for installation , purposes. Further, there is no documentation of any i inspections of the SGUL assemblies between the receiving inspections and the pre-pour inspections of the embedments. According to an interview conducted by the third party l (Reference 9.30) one QC inspector recalled that the components l_ were " assembled" on the ground and subsequently " disassembled" before installation. There was general agreement on the following installation sequence: O

l Revision: 0 Page 24 of 62 RESULTS REPORT ISAP V.b (Cont'd) L 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) [ - 18 pieces of No. 18 rebar were cadwelded to each i embedment plate (see Figures 2 and 3), L - the embedments were then hoisted into place, the rebar was cut to the length required by design, l plates for anchorage were cadwelded to the other ends [ of the No. 18 rebar, the SGUL beam was lifted and bolted into place with four bolts, and, finally concrete was poured. Following the concrete pour, the remaining bolts were installed. A millwright recalled (Reference 9.31) that "... l . there was some difficulty in installing the bolts due to misalignment of the holes in the shims with the threaded holes in the embed plates. The shim holes were enlarged by grinding away that portion of the shim that obstructed the threaded hole ... the threaded holes were cleaned ... and the bolts installed to a snug tight condition". There is no documentation as to when the remaining bolts were installed. There is also no indication of any involvement of G6H 1 engineering in the decision to enlarge the holes in the shim pack. DIR D-2481 was written to document this discrepancy. An investigation into the causes of the SGUL connection misalignment is described in Section 5.5.1. 5.3 Inspection of Thread Engagement in Other Populations of Bolted Connections To assess the potential generic applicability of the SGUL bolt I thread engagement issue to bolted connections dependent upon thread engagement, other installations were inspected on a sampling basis. Two populations with the feature of bolts being installed in blind holes, as in the SGUL connection, were inspected, i.e., Richmond inserts and drill and tap blind connections. The following sections present the results from these investigations. c ,

Revision: 0 Page 25 of 62 RESULTS REPORT ISAP V.b  ! (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) 5.3.1 Inspection of Richmond Inserts The Action Plan called for an investigation of the Richmond insert population associated with pipe support applications (both large bore and small bore) using sampling techniques consistent with Appendix D of the CPRT Program Plan. It was determined at the inception of this task that a similar, but more encompassing, investigation'was being conducted as part of the QA/QC Review Team ISAP VII.c, " Construction Reinspection / Documentation Review Plan" (Reference 9.32). It was decided that the QA/QC Review Team investigation results satisfied the objective of the investigation planned under this ISAP. The results of the QA/QC investigation are documented in Appendix 33 of the ISAP VII.c Results Report, " Concrete Insert Thread Engagement", and are discussed below. The scope of the investigation included Richmond insert applications not only for large-bore and small-bore pipe support applications, which was the scope of the original plan in ISAP V.b, but also for structural steel and pipe whip restraint applications. The ISAP VII.c investigation utilized proportional sampling to select samples from the appropriate construction work categories. The thread engagement was determined utilizing UT measurements for 323 inserts, without disengaging any bolts from the j inserts. Approximately 13% of the inspected samples had thread engagement less than the criteria requirement of twice the bolt diameter plus 1/8 inch. An analysis of the identified deviations showed that in all cases the allowable insert loading was not exceeded. However, the likelihood of an undetected construction deficiency could not be objectively assessed so that the trend was determined to be unclassified. O o

Revision: 0 Page 26 of 62 fg RESULTS REPORT 4 1 SAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) A comprehensive program to determine the adequacy of all Richmond inserts in the plant was established by the Project (Reference 9.5). If it is determined to be necessary, based on the results of this program, appropriate corrective action will be implemented. The issue of the possibility of thread damage in the Richmond inserts is addressed in Section 5.5.2. 5.3.2 Inspection of Drill and Tap Blind Hole Bolted Connections Based on review of bolted connections where thread engagement is an important attribute, the population of structural bolted connections with a drilled and tapped blind hole was selected for inspection. A sampling plan in accordance with Appendix D of the CPRT Program Plan was utilized, O(m,/ The population of drill and tap blind connections was established by review of all structural and architectural drawings. This review revealed that for a number of such bolt installations the thread engagement requirement was not specified on design drawings or the associated DCAs. A sample of bolts was selected in accordance with the Appendix D of the Program Plan for the two subpopulations: 1) the bolts with specified thread engagement and 2) the bolts for which thread engagement had not been specified in the design documents. The acceptance criterion used for both subpopulations was a thread engagement equal to the bolt diameter. This criterion is in accordance with the requirements of the current Project specifications and is applied as a standard industry practice (see Section 5.4). The details of the investigation are documented in Appendix G of STIR-CPRT-S-001 (Reference 9.5). No deviations were found within either of the two subpopulations. Therefore, with 95% confidence, the rate of occurrence of bolts with less than the required thread engagement within each of the two subpopulations is less than 5%. The issue of the possibility of thread damage in these drill and tap blind connections is addressed in Section ( 5.5.2. -

Revision: 0 Page 27 of 62 l RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) 5.4 Review of Specification of Thread Engagement Requirements and the Associated Inspection Requirements As part of the effort to determine circumstances that led to the installation of bolts with inadequate thread engagement in the SGUL connections and lack of appropriate inspection records, as well as to numerous Richmond inserts having less than required thread engagement, the process by which thread engagement requirements for bolting were specified in the design documents and the associated inspection requirements were reviewed. Conditions both as they were at the time of the SGUL installation (1978-1979) and as they are currently were reviewed. Based on the objectives stated above, the review focused on structural bolting and Richmond inserts. Throughout the period of construction of the plant, the FSAR has invoked the requirements of the AISC Specification, 7th s Edition (Reference 9.33) as a commitment applicable to steel y design and construction. The AISC Specification specifies a thread engagement requirement for structural bolting by stating that the head of the installed bolt has to be at least flush with the nut. The AISC does not specify any specific requirements for bolts that are engaged into a threaded hole, e.g., a drilled and tapped hole. From the " flush with the nut" requirement, it could be inferred that a requirement for bolts in threaded holes is an engagement length equal to the depth of a standard nut. This is equal to the bolt diameter. It is common industry practice to apply this requirement as a rule of thumb. Revision 0 of the specification 2323-SS-16B, 

                                               " Structural Steel (Category I)" was in force in 1978 and contained a section on structural bolting but did not specify specific thread engagement requirements. However, consistent with the FSAR, it invoked the AISC requirements, and thereby the " flush with the nut" requirement for thread engagement was applicable.

For structural bolts in threaded holes, the thread engagement ) requirement is typically specified on the design drawings. l While performing the task discussed in Section 5.3.2, it was j discovered that for some of the bolts, the thread engagement l requirement was not specified on the design drawings (DIR j D-2477). I O

Revision: 0 Page 28 of 62 ; RESULTS REPORT ISAP V.b (Cont'd) i 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) I ( In 1978, the QC inspection procedure CP-QCP-2.9, " Erection and Bolting of Structural Steel" specified, consistent with the AISC, the requirement of " flush with the nut", as an inspection attribute. .No criteria or inspections were specified for bolts engaged in threaded holes. As stated in the quote in Section 1.0 of this report, UT measurement of i this kind of bolts was not a procedural requirement. 1 The current specification 2323-SS-16B, Revision 1, specifies the " flush with the nut" requirement and further specifies for bolts in drilled and tapped holes that the thread engagement length has to be equal to the depth of a standard nut or as specified on design drawings. It also requires 100% inspection of thread engagement of structural bolting,  ; including UT measurement of blind bolted connections to j determine thread engagement. The current QA Procedure  ; NQA-3.09-2.01, " Inspection of Structural / Miscellaneous Steel", Revision 0, which implements the inspections specified in O 2323-SS-16B, contains this inspection requirement. A thread engagement requirement (or more accurately for Richmond inserts, requirement for insertion length) for bolts  ; or threaded rods engaged in Richmond inserts was first l specified in the Project design documents in April 1981 when the requirement of twice the bolt diameter plus 1/8 inch was  ; specified in the specification 2323-SS-30, " Structural Embedments". This requirement was not specified as a QC inspection attribute until June 1982 when it was included in the QC instruction QI-QP-11.14-8, " Verification of l Installation of Richmond Insert Bolts". It is noted that 

                                                                                                  )

installations utilizing Richmond inserts were made before the i thread engagement requirement was specified in the Project documents. Currently the specification 2323-SS-30 (Revision 3) specifies the same requirement (twice the bolt diameter plus 1/8-inch), and establishes that this attribute is to be 100% QC-inspected by UT measurements. The current QA Procedure NQA-3.09-2.05, 

                           " Civil-Concrete Anchor Inspection Activities", Revision 0, which implements the inspection requirements specified in the           j specification 2323-SS-30, includes this requirement.                     '

As discussed in Section 5.3.1, numerous findings of Richmond inserts with engagements less than the specified requirement were made in a CPRT investigation of the Richmond inserts. A f comprehensive corrective action effort to rectify the problem is underway, as explained in Section 5.3.1. 

                                                                                          ..        I

Revision: 0 Page 29 of 62 gs RESULTS REPORT (' ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) 5.5 Investigation of Thread Damage While removing the SGUL bolts during the initial inspections l of the bolts and threaded holes of the embedments (see Section i 5.1), binding of the bolts due to misalignment of the holes in the SGUL beam end plates, shims, and the embedments was observed. Thread damage was also observed in some of the embedment holes. Based on these observations, the third party . concluded that the main cause for the thread damage was  ! misalignment of the bolt holes in the SGUL beam end plate, the ' shims, and the embedments. To investigate factors that led to the SGUL misalignment and the ensuing thread damage, the applicable aspects of the fabrication and installation of the SGUL assemblies were reviewed (Reference 9.21). To determine whether similar conditions and the consequent possibility of thread damage might occur in other drill and tap blind connections, the installation process of fifteen subpopulations of drill and tap blind connections was investigated by reviewing installation documents, e.g., [)/ 

               \_-                                            travelers, QC inspection records, and procedures that were in force at the time of the installations. The effect of the geometric configuration of the installations was also considered. Further, threaded rods were removed from Richmond inserts used in voided pipe supports for visual inspection of thread condition.

5.5.1 Misalignment of SGUL Connections Examination of design drawings, vendor drawings (Reference 9.9) and certain construction records indicate that some of the matched components provided by the vendor (AFCO) are not located in the positions  ! intended by the vendor. The misalignment appears to be ' related to mismatching of the embedments, shims and beams that were intended to be match-mated as indicated on AFC0 drawings. This investigation considered the potential causes of the situation from two perspectives:

1. Fabrication and installation / fit-up of the .

support beam assemblies was possibly less I than adequate, and l l {"'g 2. Procedural controls for the installation of j s_,/ the assemblies was possibly less than adequate.

1 Revision: 0 Page 30 of 62 

           ~

RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) The investigation into the first hypothesis found that l the AFC0 drawings (approved by C6H) show that each beam was identified with a piece mark, and that each embedment plate and mating end plate were match-marked during fabrication to indicate the required orientation between the two. Further, during fabrication, each end plate was match-drilled with a particular embedment. l Based on the information about the fabrication process and on the vendor QC documentation, fabrication does not appear to be a cause of the misalignment condition. In pursuing the first hypothesis further (possibility of improper installation / fit-up of the. assembly), it was determined that there is no documentation indicating that CPSES construction personnel used the AFC0 drawings that define the location and orientation of each support beam component. Furthermore,'the information required to match up match-mated components O and the requirement to do so was not transferred from the AFC0 drawings to the G&H design drawings (DIR D-2479). Concrete. pour card packages identify the piece marks of some embedments by' hand-written annotations (References 9.34 and 9.35). It was determined from me. king a comparison between these pour card packages and the AFC0 drawings that five of the  ; embedments are in the correct location, one embedment (identified by piece mark) is located in a steam generator compartment that is not in conformance with the AFC0 drawings, and no documentation exists for the  ; remaining two e nbedments. Therefore, at'least two, and .i possibly three, embedments are in the incorrect 'l locations. This fact alone implies that alignment I should not have been as significant a problem, suggesting instead that while the embedments were properly located, the orientation of some of them may j be incorrect (e.g., the embedments may have been l rotated 180'). f Although no records could be found to document the locations of the SGUL beams, the third party determined  ; by a walkdown (Reference 9.25) that all the SGUL beams ] (and, therefore, the attached end plates) are located 1 and oriented as required by the AFC0 drawings. O Location and orientation of the shim packs to be matched with the beams and corresponding embedments could not be established during the walkdown so that no conclusion

Revision: O Page 31 of 62 I 

                           ,_                                RESULTS REPORT
                       \-                                       ISAP V b (Cont'd) i i

5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) , can be drawn with respect to whether proper matings of end plate-to-shim pack-to-embedment were maintained. Likewise, orientation of the embedments could not be verified from either the pour card packages or from the third-party walkdown. The investigation of this first hypothesis indicates that one probable cause of the misalignment condition was the failure to assemble the match-mated SGUL beam end plate, shim pack, and embedment components in accordance with the match mating information provided on the vendor drawing. It is believed that this was due to the fact that the vendor drawings were not used for construction and the match mating information was not transferred to the drawings that were used for construction (DIR D-2479). However, the fact that all four SGUL beams, and five of eight embedments, were in correct locations suggests that there was some (} 

                       \s ,/

knowledge of the locations of the SGUL beams specified in the vendor drawings. The investigation of the second hypothesis (i.e., possibility of inadequate procedural controls for the installation process) consisted of a review of the procedures and instructions related to miscellaneous steel installations (Reference 9.21). This review identified four discrepancies:

1. There existed a lack of traceability requirements for certain miscellaneous steel items in that the controlling procedure (CCP-31) was limited to items that were l embedded and did not extend to attachments I such as the shim packs and SGUL beams.

Accordingly, control for certain SGUL components was lost following receiving I inspection. 1

2. Installation procedures related to the SGULs i lack the speciatcity needed to assure proper I assembly (particularly in regard to the I matching of individual components during assembly). There are no provisions in the installation procedure (CCP-31) detailing the
                             ~

matching requirements and there is no - reference to the vendor drawings that contain I this matching information.

(l 9 Revision: 0 4 Page 32 of 62 RESULTS REPORT ISAP V.b I (Cont'd)  ! i1 5.0 ~1 IMPLEMENTATION OF ACTION PLAN'AND DISCUSSION OF RESULTS (Cont'd) j 1

3. QC documentation could not be located for l

l inspections that were required by one of the procedures (CP-QCP-2.17). 

                                                                                                                     ] i
4. A OC inspection procedure (CP-QCI-2.17-1) was .,

not properly implemented such that inspection records were not generated for two of the ]- 4 SCUL embedments. DIR D-2480 was written to document concerns 1 and 2 1 above. DIR D-2485 was written to document concerns 3 and 4 above. The results of the investigation of the second hypothesis indicates that a lack of procedural controls must also be considered a potential cause for j the misalignment condition. 5.5.2 Investigation of the Possibility of Thread Damage in Richmond Inserts and in Drill and Tap Blind Connections The possible generic applicability of the thread damage problem identified in the SGUL connections was assessed by performing an investigation consisting of two tasks:

1) threaded rods from Richmond inserts were removed from voided pipe supports and inspected for thread condition, and 2) the installation process of fifteen subpopulations of drill and tap blind connections and the associated documentation were reviewed. Neither investigation was based on sampling. Details of the I investigation are provided in Appendices "L" (Richmond Inserts) and "I" (Drill and Tap Blind Connections) of STIR-CPRT-S-001 (Reference 9.5).

Appendix L " Thread Condition of Bolts / Rods in Richmond Inserts", contains the procedure for inspection of the thread condition of the threaded rods in the Richmond inserts used on pipe supports that have been voided as part of the large-bore piping requalification program. The population of applicable supports was determined i and the supports then disassembled to observe the condition of the threaded rods used in the Richmond inserts. Eighteen rods were examined; none exhibited thread damage that would reduce the load-carrying l capacity over the length of the required thread .j engagement (all rods were found to 1

Revision: 0 Page 33 of 62 RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) exceed the minimum thread engagement requirement of two bolt diameters plus 1/8 inch). Based on these results, the fasteners in Richmond inserts are not prone to thread damage. The third party concurs with this conclusion. Appendix I, " Procedure for Evaluation of Potential Thread Damage Due to Misalignment" addresses the other i population germane to this investigation, namely drill and tap blind connections. SWEC performed an evaluation of this population, the results of which are presented in Appendix I (Revision 0, dated July 15, 1987) of STIR-CPRT-S-001. The third party reviewed this appendix (Reference 9.36), and the results of the third party review are presented here. The SWEC evaluation criteria were established based upon the third-party investigation of. installation / fit f up of the SGULs. The factors identified by the third party and input to the SWEC investigation were: lack of awareness of construction personnel of necessary l installation drawings, specifications, and procedures; lack of inspection plans and inspection reports; . match-drilled, multi-hole assemblies that were shipped separately and could be interchanged during installation; and improper grinding of bolts in conjunction with authorized bolt cutting. The total population was subdivided into fifteen  ; subpopulations that represent installations that share similarities in design or equipment type. These . subpopulations were CRDM cables, electrical equipment, single bolt attachments, radiation monitor, 480 V switchgear, missile barriers, moment-limiting restraints, stairs, environmental barriers, pipe bumpers, lead shielding frames, pipe whip restraints, equipment not installed, non-nuclear safety-related, and reference drawings. The evaluation of each subpopulation was conducted by determining whether there are any similarities between the subpopulation documentation and equipment requirements and those of the SGULs as identified in the evaluation criteria. Several subpopulations were not reviewed in detail because generic considerations preclude the possibility of thread damage or improper

Revision: 0 Page 34 of 62 

      . f-w                                                RESULTS REPORT k

ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) grinding of bolts. It was determined that blind hole details in the stairs subpopulation were not used in the installations. The subpopulation labelled 

                                             " equipment not installed" included items installed after June 24, 1986 when existing procedures and inspection requirements were adequate to prevent misalignment and thread damage. The items included in the non-nuclear safety-related subpopulation were excluded from the evaluation. Finally, the " reference drawing" subpopulation included those drawings referenced on DCAs but where the DCA is accounted for on another drawing.

Evaluation of the remaining subpopulations revealed a number of significant differences between the other blind hole installations and the anchor bolt installations of the SGULs. Almost all subpopulations included installations for which construction travelers f} 

          \~/

were issued (single bolt attachments and bolted connections in pipe whip restraints were not installed in accordance with travelers). The significance of travelers is that each consists of a detailed procedure established for performing the installation, based upon design considerations, and the requirements for QC witness, inspection and verification at appropriate points in the procedure. The third party review of Appendix I (Reference 9.36) included an assessment of all QC procedures and instructions referenced on the travelers relating to I bolt installation to determine whether or not sufficient control was exercised to provide confidence l that proper installation occurred. This review l established that the attributes identified in the procedures and instructions referenced on the travelers were, in all cases, sufficient to provide confidence that the installations were properly performed. In most subpopulations, installation involved either standard-sized bolts or bolts supplied with the items so that bolt cutting was unnecessary. The third party also noted that bolt cutting requires that personnel obtain equipment that would not ordinarily be used ("'g during installation of items with standErd bolts. N.)

_-_-_---_-_-_--_--_------------------------------------------------------------------------------T Revision: O Page 35 of 62 RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEME'NTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) The third party further noted that there are significant differences between the equipment i characteristics of the SGUL assemblies and those of all i other blind hole installations, such that the former l represents a unique item regarding the possibility of misalignment of holes. The SGUL assemblies are unique in that they consist of very large and heavy pieces of steel that are difficult to position for fit-up; include embedments into which the bolts were installed; consist of fabricated pieces that were match-drilled to close tolerances in specific one-to-one matings; consist of non-adjustable assemblies that connect two embedments in fixed locations; and require the alignment of several holes for the installation of each bolt (end plate, several shims and embedment), for numerous bolts in each end plate, and for two end plates - all simultaneously and in fixed relational locations. O Based upon the significant equipment characteristic, installation, inspection and documentation differences identified above between the other blind hole installations and the SGULs, the third party concurs with SREC's conclusion that it is unlikely that thread damage has occurred in other installations. The third party also concludes that it is unlikely that there was unauthorized grinding of bolt threads in other blind hole installations. 5.6 Discussion of DIRs The following provides a discussion of DIRs generated during the implementation of ISAP V.b and other DIRs relevant to this ISAP. 5.6.1 Construction-Reinted DIRs The construction related DIRs document concerns identified in three areas - Hardware, Installation / Procedural Control and QC Inspection Requirements and Inspections. O

Revision: O Page 36 of 62 k RESULTS REPORT [ 

 \~                                                      ISAP V.b                                                                          '

(Cont'd) 1 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) Hardware DIR E-990, E-1055, E-1056, and E-1224 document concerns expressed by external sources regarding the engagement adequacy of the SCUL bolts. DIR E-1055 and E-1055 also identify the concern that this lack of adequate l engagement may be the result of the bolts having been l cut. DIR E-990 also expresses concerns regarding the ' potential generic applicability of the thread engagement problem. Because DIRs E-990, E-1056 and E-1224 represent a violation of a design commitment, i.e., providing a thread engagement that resulted in the allowable stresses being exceeded, each is a deviation. DIR E-1055 was determined to be k unsubstantiated because the alleged cause of bolt cutting was determined not to be related to the presence of debris in the hcit hole, and is therefore closed. DIRr E-990, E-1055 and E-1224 were clased. l /T based on reanalysis and rensaembly of the SGUL kj connections using new bolts. The generic applicability portion of DIR E-990 is closed because of the corrective action underway to resolve the Richmond insert thread engagement issue, se discussed earlier in this Results Report. DIR D-2476 documents the finding of 1) thread damage in some of the SGUL embedment holes and 2) grinding of 1 the bolt threads at the end of some of the bolts. This discrepancy is classified as a deviation and is closed on the baals of the corrective actione discussed in this Results feport. Installation / Procedural Control DIR D-2480 documented Isck of crecific procedural instructions for the installation of the SGUL components. These procedural instructions were required because specific mating and matching considerations as well as specified thread engagement were important to ptoper installation. The procedure also lacked requirements for material traceability for such components. This DTR was classified as a [ deviation and is closed on the basis of project 

   -~g                                 corrective action performed under Correcciye Action Report (CAR)-101.
 %.)J                                                                                                                                    ;

_____u-.--_ - , - - - - - - -

Revision: 0 Page 37 of 62 

            ~'

RESULTS REPORT ISAP V.b (Cont'd) j 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) DIR D-2481 documented the failure of construction personnel to identify arid report problems encountered during SGUL installation to engineering. It also identified the circumstances wherein construction personnel performed grinding of bolt threads, shim packs and end plates to facilitate insertion of bolts into misa11gned holes without engineering approvals. This unauthorized modification of Seismic Category I hardware is classified as a deviation. Project activities under CAR-101 also allow this DIR to be closed. QC Inspection Requirements and Inspections DIR D-2482 documented the leck of QC inspection requirements in the QC procedures and instructions for the inspection of thread engagement of bolts installed in drill and tsp blind connections, such as the SGUL [,,_ bolts, and of bolts or threaded rods in Richmond 

      '~-                                                  inserts. This DIR was classified as a deviation and is closed as a result of improvements made to procedures, as discussed in Section 5.4..

DIR D-2483 documented QC's failure to $dentify and report the problems encountered during installation of the SCULs. QC should have identified the misalignment problems experienced as well as the modifications (grinding) performed on the bolts and other parts of the SGULs. This DIR was classified as a deviation, and transferred to the QA/QC review team for their consideration. DIR D-2485 documented the concern that certain QC documentation that should have been generated in accordance with CP-QCP-2.17 could not be located. It alsc identified that for two SGUL cabedments, material traceability information was not occumented on a i checklist as required by CP-QCI-2.17-1. Thic DIR was classified as a deviation, and transferred to the QA/QC review team for their consideration. 5.6.2 Design-Related DIRs ( DIR E-1208 docucents external source concerns about the N. design of SGUL connections. This DIR was merged with DIR D-2484 because that DIR expresses the same concerns more specifically.

Revision: O Page 38 of 62 RESULTS REPORT ISAP V.b 1 (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) DIRs E-1209 and E-1211 documented concerns about the adequacy of the design calculations for the anchorage l of the embedments to the concrete structures. These DIRs were merged with DIR D-0003, which expressed the concern about the adequacy of design calculations in more detail. DIR D-0003 was generated within DSAP VIII* and documented the failure to account for the bending moments in the design calculations for the SGUL embedment anchorage. This was consistent with the assumption of the original design concept that the SCUL connections were a simply supported beam and this only transmitted beam and thus only transmitted shear, not ' bending moments. However, the actual connection detail did not provide a pinned condition but rather a moment resisting rigid connection. The design calculations for the SGUL beam and the embedment anchorage were also O based on this incor rect assumption. This led to a conservative design of the SGUL beam, but an unconservative design of the embedment anchorage. i DIR D-0003 also expressed the concern that bounding LOCA loads were not used in the steam generator compartment analyses. These analyses provided the load information for the evaluation of SGUL connections. DIR D-2484 documented the lack of design calculations to size the SGUL bolts and to establish the thread engagement requirements. This failure to provide design calculations for components of the load path of the SGUL that is part of a Seismic Category I structure is a violation of the design requirements, and ' therefore is a deviation. This DIR was closed cn the basis that the reanalysis of the SGUL and the hardware modifications corrected the identified concerns. DIR D-2477 documented the lack of thread engagement requirements in specifications for blind hole bolted connections and Richmond inserts, as well as the lack of specification of thread engagement requirements in structural design drawings for a number of drill and tap blind connections, including the SGULs. It is 

                )
  • DIRs identified by the Design Adequacy Program (DAP) will be closed in accordance with the r' requirements of the CPRT Program Plan, Rev. 4 i

l I

Revision: O Page 39 of 62 RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) common industry practice to specify thread engagement requirements for bolted connections such that the full tensile capacity of a bolt can'Le developed. The lack of specification of thread engagement requirements in the Project documents is a violation of design requirements and is, therefore, a deviation. The current gecifications, drawings, and procedures are adequate, and therefore this DIR is closed. DIR D-2479 documented engineering's failure to incorporate information required to ensure proper mating of SGUL components into the drawings or specifications provided to construction. This information was contained on vendor fabrication drawings. Failure to provide this relevant information to allow proper installation is a deviation. This DIR is closed on the basis that the SGUL has been redesigned and reinstalled in accordance with the new design. DIR D-2152 documented the concern that certain DCAs that were generated during the early implementation of this ISAP concerning thread damage and enlargement of SGUL connection holes were not considered. Another issue raised by this DIR was a possibility of lamellar tearing of the embedments and the corresponding reduction in pull-out capacity. The AISC specification that is applicable to the embedment dc,es not require reduction of capacity due to lamellar tearing considerations. This issue was classified and closed as an observation. DIR D-2478 expressed the concern that the design drawings did not specify a maximum thickness for the shim pack of the SGUL connection. The thickness of the l shim pack in a bearing-type connection (as provided in l the original SGUL connection design) affects the i bending stresses in the SGUL bolts. The failure to j' specify the maximum acceptable shim pack thickness is a l deviation. This DIR was closed on the basis that the i SGUL connection was redesigned and the redesign has l been implemented in the field. i O 1

Revision: 0 Page 40 of 62 RESULTS REPORT ISAP V.b (Cont'd) 

                                                                                               ]

1 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) 5.6.3 Design-Construction Interface DIR D-0017 documented a failure to incorporate a revision of SGUL loads by Westinghouse to G&H design calculations. This potential problem in the interface between Westinghouse and G&H was transferred to the l DSAP VIII as a concern that would be more appropriately I tracked and evaluated sithin that DSAP. As discussed in Section 5.6.2, DIR D-2479 was written to document the failure to incorporate information provided by the fabricator into drawings and specifications provided to construction. Since this failure could also be considered an interface problem, it is identified in this section as well. 5.7 Safety Significance Evaluation hv Of the nineteen DIRs written under this ISAP, one (D-0017) was transferred to DSAP VIII; one (E-1055) was closed as 

                          . unsubstantiated; one (D-2152) was closed as an observation.

DIR E-1208 was merged with DIR D-2484. Two DIRs (E-1209 and E-1211) related to design calculations evalueting the pull-out capacity of the SGUL embedments were nerged with DIR D-0003. Although DIR D-003 was identified by the DAP and will be closed in accordance with Revision 4 of the Program Plan, the analyses performed in response.to ISAP V.b and DSAP VIII concerns ( Reference 9.16) demonstrated that the SGUL would not lose its capability to serve its safety function as a result of a postulated seismic event and a concurrent LOCA. Therefore this issue was not safety si Fnificant. All of the remaining DIRs involved activities which either identified improperly installed SGULs or other components or identified activities which contributed to their improper installation. Analyses to deter 1nine whether or not the SGULs with the originally installed bolts would have been able to perform their required safety function were not performed. Because of the extensive corrective actions instituted by the l Proj ect , it was determined to be more expeditious to proceed l directly to an analysis of the root cause and gencric 1 implications without evaluating the safety significance of the 

           - N             deviations. Therefore the remaining E-type DIRs (E-0990, E-1056, and E-1224) and D-type DIRs (D-2476 through D-2485) were classified as unclassified deviations. Table 1 summaries the DIRs and their respective classifications.

l l

Revision: 0 Page 41 of 62' RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) 5.8 Root Cause and Generic Implications The design requirements for the. support systems for the reactor coolant system components, including the steam generators, require that these components be restrained such that their integrity is not compromised in the event of design basis events such as an SSE or a LOCA. In order co provide for this support, steam generators have typically been-provided with a supporting structure that resists vertical loads (e.g., der.dweight, seismic, LOCA, etc) and a structure / structures that resist loads in the horizontal plane (e.g., seismic, LOCA, steam line break, etc). The design of the SGUL, one of the supports that resists loads in the horizontal plane, was the responsibility of G6H. In its design, G6H used loads that were provided by Westinkhouse. The process leading from design to installation of such safety-related, Seismic Category I components would typically involve establishment of the design concept, completion of the O design (including calculations, drawings, etc), specification of requirements for the hardware, identification of installation requirements and installation and inspection of the hardware. As discussed in Section 5.6, the unclassified deviations fall into the following general categories and subcategories:

1. Construction Hardware Installation / Procedural Control QC Inspection Requirements and Inspections
2. Design (

1

3. Design Construction Interface The initial concern about alleged unauthorized modification of the SGUL bolts ultimately manifested itself as a broader problem associated with the design and installation of the i SGULs and included facets from each of these categories. The I

() root cause evaluation discussed below is conducted along the lines of these categories. ,

Revision: 0 , Page 42 of 62 ' f RESULTS REPORT t ISAP V.b (Cont'd) 5.0- IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) Construction-Related Problems Several causes can be postulated for the observed hardware deviations:

1. Less-than-adequate installation procedures and instructions (or complete lack thereof) arising from:

Less-than-adequare specification of installation and inspection requirements by engineering and/or, Less-than-adequate incorporation of installation and inspection requirements-in procedures by construction

2. Less-than-adequate performance by construction craft and supervision
3. Less-than-adequate inspection procedures
4. Less-than-adequate performance of inspections to the requirements of the inspection procedures Installation and inspection activities related to safety-related systems, components and structures are to be conducted to procedures and instructions that contain sufficient detail to ensure that the design intent is met and that installation activities are adequately controlled such that a quality product results. Accordingly, provisions embodied in the procedures are a dual responsibility of design and construction organi:ations. In the case of the SCUls, G&H had the responsibility to convey design requirements in the form of drawings and specifications and Brown & Root had the responsibility to convert these. requirements'into {

implementable procedures, including provisions for quality ' control inspection of'the installation process. G&H design drawings identified the minimum thread length of the tapped holes within the embedment plates. Specific j information addressing the associated engagement requirements for the bolts was not documented on the drawings; however, in I the absence of such information, it is judged to be logical to conclude (given the " min." notation on the drawing) that it O was the designer's intent that the bolts were to be engaged to at least this " min." dimension. In I L_______i____________ - - -

l Revision: O Page 43 of 62 RESULTS REPORT ISAP V.b (Cont'd) I 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) practice (as discussed below) unless the holes were tapped greater than the " min" dimension, it is difficult to obtain engagement to this dimension. Had construction actually attained this degree of engagement without encountering other L problems such as thread damage, the SGULs would have been l acceptably installed per the original design intent. (Note that the following section on design-related problems identifies problems with the original design.) Attainment of this minimum dimension (engagement) would have been difficult, independent of alignment problems, given the physical tolerance constraints posed by the geometry of the overall SGUL assembly and the location of its attachment points. This aspect of the design and the associated challenges of fabrication and installation most probably contributed to a greater extent to the inadequate engagements than the lack of specificity of the design drawings (i.e., working to close tolerances within an array of bolts requires O greater precautions and levels of control). Given the special design considerations of the SGULs and the challenges of installation, it is reasonable to expect that the designer would have provided some form of guidance to j construction, by outlining controls for fabrication and i installation that were important to attaining the design objectives. Evidence was not fcund that this was done. In the absence of fabrication and installation guidance from the designer, the constructor and fabricator should have recognized these installation challenges and either resolved them internally or addressed them with the designer. It is apparent that AFCO, the fabricator, understood the critical nature of the SGUL assembly and its special requirements for fabrication / installation and incorporated controls such as match-mating and traceability of parts. However, this information was either not properly conveyed to construction, or construction failed to incorporate the information into procedures. G&H approved the AFC0 fabrication drawings, which contained details of match-mating and parts traceability; however, G6H did not update its design drawings to reflect this information. While it has been determined that these O non-updated G&H drawings were used in the field, it is not clear the extent to which this contributed to the alignment

v. I Revision: 0 Page 44 of 62 RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) problems associated with mismatched parts. Also it is unknown whether construct;>n had access to the AFC0 fabrication drawings such that they could have properly incorporated this information. It is clear, based upon knowledge of the fabrication process, that if construction had access to the detailed parts matching and orientation requirements that were specified on the AFC0 drawings and had construction used this information er guide the location and orientation of the embedments with their respective match-mated beams, the misalignment problems could have been avoided. Based upon the partial success that was attained in getting the correct parts to location, it appears either construction had the AFC0 drawings or related information, or that a best-effort attempt was made by construction to match and mate parts. However, full success in terms of attaining correct alignment-also requires having the beam end plate-embedment plate orientation exactly the same as existed in the fabrication shop at the time the plates were match-drilled. This did not occur. Accepted industry construction practicen would have included the preparation and use of specific ir- 'llation procedures or instructions for challenging hardware Installation such as the SGUL installation. The only procedure that was available for use was a very general procedure CCP-31 for " Installation of Miscellaneous Steel". Thus, construction failed to develop a set of adequately detailed procedures thet would have conveyed the installation guidances that had been made available by G6H, and that would have represented good construction practice wherever the G&H-provided guidance was not specific. Collectively, the fact that installation procedures and guidance were less-than-adequate, due both to G6H's failure to provide appropriately detailed guidance and to construction's failure to develop specific proceduralized installation guidance, is a root cause of the SGUL installation difficulties. Both G&H and construction share the responsibility for either not recognizing the special challenges of the SGUL installation, or for not taking action, either individually or jointly, to assure success. O

l l Revision: 0 Page 45 of 62 

           )                                         RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd)

With regard to a second postulated cause of the hardware deviations, i.e., less-than-adequate performance by construction craft and supervision, there are a range of installation activities that craft and supervision are expected to be capable of completing without specific procedural guidance. Likewise, there are certain situations with which craft should never have to deal without adequate procedural guidance. Had installation instructions for the SGUL beams and their match-mated embedments and shim packs l been properly specified, including installation sequencing requirements, all components could have been properly mated and oriented, and the craft would have had to deal only with standard bolt installation considerations such as thread engagement, torque sequence, and torque requirements. 1 In the absence of any thread engagement requirement, craft should have resorted to the rule of thumb of " engagement ($ should equal at least one bolt diameter". Although full 

 \                             compliance with this rule of thumb would not have been achievable in this instance, it is evident that this " rule" was either not understood or ignored due to the difficulties of installation with misaligned parts.

In the case of the SGUL bolts, the craft chose to proceed with installing the bolts by resorting to such obviously unacceptable practices as grinding the holes, grinding off the crests of the threads on the ends of the bolts, forcing the bolts into the embedment holes with such excessive torque that the threads in the hole were damaged, and cutting some bolts so short that it must have been obvious that the bolts as installed had achieved only a fraction of their intended engagement. Such practices by the craft are unacceptable. When difficulties were encountered, supervisors should have been informed and, as appropriate, the problems should have been elevated to construction management. Accordingly, it is , concluded that the craft and/or supervision failed to perform 4 adequately. This failure of craf t and their supervision to l perform adequately was also a root cause of the SGUL 

                    ,         installation problems in that they 1) resorted to practices 5

that were clearly unacceptable even in the absence of 1 procedural guidance and 2) may have failed to inform their management of obvious problems. 1 1 

                                                                                                 ----_______--_n

Revision: 10 Page 46 of 62 RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) The third postulated cause for the hardware deviations is less-than-adequate inspection procedures. While inspection procedures historically addressed bolted connections with nuts, drill and tap blind bolted connections were not addressed with respect to any provisions to verify engagement. Accordingly, the failure to provide.for and implement such inspection requirements eliminated the opportunity to identify engagement deviations such'as those associated with the SCULs. Therefore, this failure to develop adequate inspection procedures is also a root cause. The fourth postulated cause for the hardware deviations is less-than-sdequate performance of inspections. The failure to inspect the SGUL and its connections, a safety-related assembly, represents a violation of 10CFF50, Appendix B quality assurance requirements. The third party review was unable to establish whether or not the SGUL connections were inspected and if they were, why documentation of such inspections was not found. A potential exists that the NCR-issue related to the question of whether or not the SGUls were ASME Section III, NF supports may have confused QC personnel. This cannot be confirmed. (Inspections of NF supports are the responsibility of Brown & Root, holder of the "N" stamp, while other supports are inspected by TUGCO.) Had proper inspections taken place, the problems with the SGUL bolt engagements would have been identified. Thus, the fourth postulated root cause ir partially substantiated on the basis that documentation of SGUL inspections could not be located by the third party. The failure to identify engagement problems with Richmond inserts during the mid-1982 inspections (reference ISAP VII.c Results Report, Appendix 33) indicates that these inspections l' were inadequate. Had a proper inspection and documentation taken place, the problems with the Richmond inserts engagement would have been identified. Therefore, the lack of QC inspections and failure to implement QC inspections adequately is also a root cause of the problem with Richmond insert thread engagement. 

                                                                                                                                     ~

Revision: O Page 47 of 62 RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) Design-Related Problems The postulated causes of the design-related problems are less-than-adequate design control and lack of experierce of the design organization. Design-related discrepancies were identified in the areas of design calculations and development of drawings and specifications and in the interface between design organizations (G6H and Westinghouse). Regarding problems with Jesign calculations, the original design concept for the SGULs was that they would behave as simple pinned beams. Such a design would have resulted in the connections between the beam end plate and embedment transmitting primarily shear loads. However, the design that was detailed by G6H as shown on design drawings v i behave as a rigid connection that would also result it transmittal of bending moments and axial forces throagh the connection. Consideration of these previously neglected forces, coupled with increased external load requirements (i.e., Westinghouse G revision of loads and errors in selection of the controlling load case) ultimately led to the redesign of the connection from a bearing connection to a friction connection. This was required in order to satisfy bolt allowables when subjected to all load components. In view of this decision to modify the l connection design, a safety significance evaluation of the ' ability of the original design to perform its intended safety j function was not performed. The calculations that were completed for the embedment were l not adequate in that errors were made in the selection of the controlling load case for design, and an erroneous assumption was made regarding credit that can be taken for shear friction acsociated with embedment pt.11out. This was compounded by the l fact that loads at the connection and embedment were revised, leading to a deviation from ACI code requirements due to predicted cracking of the concrete in the vicinity of embedment. This deviation was determined not to be safety significant because the cracking does not affect the ability of the SGUL to perform its intended safety function. Finally, the shim packs picced between the beam end plate and O

Revision: 0 Page 48 of 62 RESULTS REPORT ISAP V.b (Cont'd) I 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) I embedments to make up for construction tolerances were sufficiently thick.(due to a need to fill the gap) that the SGUL bolts would have been subjected to unanalyzed bending moments. Accordingly, pre-evaluated limits were not specified on the maximum shim plate thickness. Controlled calculations for sizing the SGUL bolts and their engagement lengths could not be located, although G6H engineers recall that they were created. The inadequacies in the original design concept, missing calculations, and errors in the design calculations represent less-than-adequate design control and are contributing causes to the inadequate bolt engagement problem. G6H failed to convey the SGUL design intent adequately with respect to bolt engagement and placement of limits on items important to the design, such as maximum shim pack thickness. g Additionally, G&H failed to provide guidance on the sequence 

           )

of assembly of the SGUls and notes of caution to assure ease of fabrication / installation in conformance with the design intent. Normally, designers elect to document such guidance on drawings and specifications. The specificity is usually governed by the complexity of the design and/or constraints to be encountered in the field. Less complicated designs that are encountered by construction typically require less specific guidance. Conversely, challenging installations such as the SGULs require more guidance than was developed by G&H. Accordingly, the failure to develop this guidance within drawings and specifications represents less-than-adequate design control and is a root cause of the SCUL bolt engagement problem. Regarding problems with the interface between G&H and Westinghouse, the updated LOCA loads that were issued by Westinghouse in 1984 were not incorporated into the G6H analysis of the steam generator compartments. Westinghouse transmitted the revised loads; however, G&H has no record of receipt. This represents a breakdown in the receipt acknowledgement process between the two organizations. As discussed, this problem exacerbated the overall problems with the design and installation deviations since the loads were significantly higher and therefore, systematically affected _O all SGUL hardware. In view of the potential ramifications of such changes, mechanisms beyond normal receipt acknowledgement

Revision: 0 Page 49 of 62 

            ,m (w_ -)                                                       RESULTS REPORT ISAP V,b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) procedures apparently did not exist to alert responsible parties. This problem is similar in nature to other interface problems identified by the CPRT Design Adequacy Program (DAP).

This breakdown, leading to the failure to convey important design input information, represents less-than-adequate design control and a contributing cause of the inadequate SGUL bolt engagement problem. The multiplicity of design-related problems is symptomatic of an overall lack of experience on the part of the engineering , organization. It is apparent that G&H either did not recognize issues associated with the constructability of the SGUL design or failed to control design activities adequately. Accordingly, it is concluded that significant opportunities for improvement of the G6H design control program existed. This conclusion is similar in nature to other findings by the CPRT DAP. The less-than-adequate design control associated f with the SGULs is a root cause of the inadequate SGUL bolt 

          \                                      engagement problem.

Design Construction Interface Problems As discussed above, there were numerous opportunities to icprove either design or construction-related programs. At the same time, the design and construction programs are interdependent with many important interfaces. These interfaces represent potential challenges to quality as well as opportunities for engineering and construction to work together through a system of checks and balances to assure quality. Thus, had the design construction interface been stronger, the difficulties arising out of initial weaknesses might have been detected and corrective action taken to resolve the specific issue, as well as to address the broader programmatic weaknesses. Summary The identified root causes of the SGUL bolt installation problem are as follows: (

1. Less-than-adequate installation procedures and I

instructions due both to the failure of G&H to provide f- drawings and specification that contained sufficiently (s) detailed instructions to guide installation, and to the l i

Revision: 0 Page 50 of 62

/

RESULTS REPORT D) . ISAP V.b (Cont'd) 5.0 1 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) i failure of construction to develop specific proceduralized guidance for the installation of the SGULs, given the information that they had been provided;

2. Less-than-adequate performance by craft and/or their supervision in that they resorted to practices that were clearly unacceptable even in the absence of specific procedural guidance and that they appeared to have failed to inform their management of obvious problems;
3. Less-than-adequate inspection procedures in that the applicable procedure contained no requirement to verify engagement of bolts in drill and tap blind connections such as were used in the SGUL connections;
4. Less-than-adequate performance of inspections, as
 -O                                                      evidenced by the fact that documentation of SGUL inspections could not be located; and
5. Less-than-adequate design control program in G6H that permitted the design drawings and specifications to be issued without specific guidance being provided for the installation of the SGULs.

Contributing causes were also identified as follows:

1. Less-than-adequate design control program in G6H that resulted in an inadequate original design concept for the SGUL connections and permitted essential calculations to be missing and to be completed in error, and
2. Less-than-adequate design control at the G&H/ Westinghouse interface that resulted in important design information not being incorporated into design calculations.

The issues identified in this ISAP extended well beyond those expressed by external sources. The engineering and construction organizations share responsibility for failing to address adequately the unique challenges of design and 

 .(                                              installation of the SGULs. The critical error was the loss of

Revision: 0 Page 51 of 62 RESULTS REPORT ISAP V.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) control of match-mating of parts. This was compounded by unorthodox and ultimately unsuccessful attempts to ameliorate the problem physically and by a breakdown in the inspection process that eliminated opportunities to identify and correct the problem. The design-related problems are similar in nature to other problems identified by the CPRT DAP. Programmatic weaknesses were identified in areas of design control, construction / installation and QC, Collectively these findings represent a failure to comply with quality assurance requirements of 10CFR50, Appendix B. The generic imolicationa of the SGUL bolt installation problem in the areas of construction and QA/QC as they relate to other bolted connections have been addressed in this Results Report and, for Richmond inserts, in Appendix 33 of the ISAP VII.c Results Report. It was concluded, based on initial evaluations of the SGUL bolt installation problem, that other p blind hole bolted connections might be similarly affected by whatever caused the SGUL problems. It wau determined through reinspection and document reviews conducted under the auspices of ISAP VII.c that there was a problem with thread engagement being less than required in Richmond inserts. An unclassified trend was identified; a corrective action program that, when completed, will correct any hardware problems associated with Richmond insert thread engagement that are identified. Any procedural inadequacies will be corrected as well. The other population of blind hole bolted connections that was implicated by the SGUL bolt installation problem was that of drill and tap blind connections. Reinspection conducted under the auspices of this ISAP determined that reasonable assurance existed that proper thiead engagement had been achieved for these connections. Specifications and inspection procedures that address drill and tap blind connections now contain appropriate requirements. No further investigations were conducted to determine the l extent of the design control program weaknesses at G6H or the implications of the causes of these weaknesses. However, as l noted above, they are similar to findings against the design control program identified during the conduct of the DAP. DAP-identified issues, as well as the other design-related issues such as the SGUL bolt installation issues, are subject to remedial actions under the comprehensive Corrective Action Program (CAP) being conducted by TU Electric. The CAP has (v) sufficient breadth and depth to address the root causes and _

l Revision: 0 Page 52 of 62 l 

  ;                                                RESULTS REPORT ISAP V.b (Cont'd) 5.0 1 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) l
                                                                                                                        )

I generic implications of the design control program weaknesses.  ! The results of the investigation in the area of construction and QA/QC have also been provided to the QA/QC Review Team for their consideration in the collective evaluation of the j l quality of construction. Accordingly, no further corrective l action under this ISAP is warranted. l l

6.0 CONCLUSION

l i The reviews of the design documents revealed that thread engagement requirements for the SGUL bolts had not been adequately specified in the design. Analyses demonstrated that the originally installed bolts and the embedment pullout did not meet the design criteria requirements. Removal of the SGUL bolts during inspections revealed inadequate bolt engagement as well as alignment problems in the bolt holes of the SCUL beam end plate, the shims, and the (~ embedment. Thread damage that was attributed to the alignment ( problem was observed in several embedment holes. The third party identified discrepancies in the load information of the steam generator compartment finite element analysis that generated the SGUL beam end loads used in the evaluation of the SCUL connections. This led to significant reanalysis of the steam generator compartments by G6H and Westinghouse. Based on the results of these analyses, Westinghouse revised the design of the SCUL connections. This resolved the original issue of the SGUL bolts. In addition, the SGUL connections are being reassembled in accordance with the revised design. To determine the potential generic applicability of the thread engagement problem, two populations of blind structural connections were inspected for thread engagement adequacy using sampling techniques: Richmond inserts and drill and tap blind connections. An unclassified trend of Richmond Inserts with inadequate thread engagement was identified. Corrective action to solve the problem has been taken by the Project. The corrective action will be overviewed by the third party as an SRT-directed activity. No discrepancies were identified in the measurements of the thread engagement of the sample of bolts in the drilled and tap blind Connections. d E__________._______

Revision: 0 Page 53 of 62 RESULTS REPORT [J.] % j ISAP V.b (Cont'd)

6.0 CONCLUSION

(Cont'd) The problem in the alignment of the bolt holes in the SGUL beam end ( plate, the' shims, and the embedment, which was attributed as the cause for the observed thread damage, was concluded to be. unique to l the SGUL installation. This conclusion was based on review of the fabrication and installation history of the SGULs and of documentation of installation 'of fif teen subpopulations of drill and tap blind connections. Documents reviewed included travelers, 1 QC inspection records, and procedures thst were in force at the I time of the installations. Implications of the geometric I configuration of the installations was also considered. Further, 18 threaded rods u re removed from Richmond inserts in voided pipe i supports to inspect the thread condition. No thread damage was I observed. Program weaknesses were evident in the design control, construction / installation and QC processes. TU Electric is implementing a corrective action program that is directed at the resciution of the issues that are potentially generic to those q O. identified by this ISAP, as well as, the correction of the root causes of past programmatic weaknesses. The Project CAP will be evaluated by the TU Electric Quality Assurance Technical Audit Program e.nd overviewed by the third-party, as directed by the CPRT Senior Review Team. ) 7.0 ONGOING ACTIVITIES The installation of the new shims in Unit 1 SGULs is ongoing. This will be followed by installation and pretensioning of the bolts, which completes the activities for the resolution of Unit 1 SGUL issues. The inspections of the Unit 2 SGULs have been initiated. The early observations indicate that same type of problems that were encountered in Unit 1 SGUL connections are also present in Unit 2, i.e., bolts with inadequate thread engagement, misalignment of bolt holes in the SGUL beam end plate and the embedment, and thread damage (Reference 9.10). The corrective action to resolve the problem of inadequate thread engagements in Richmond Inserts, in both Unit I and 2, is ongoing and is being overviewed by the third party as an SRT-directed activity in accordance with the Append 1x H of the CPRT Program O Plan.

Revision: 0 Page 54 of 62

RESULTS REPORT ISAP V.b (Cont'd) 8.0 ACTION TO PRECLUDE OCCURRENCE IN THE FUTURE
                              .No further actions are required.

9.0 REFERENCES

l 9.1 USNRC Safety Evaluation Report,-Related to the Operation of l Comanche Peak Steam Electric Station Units 1 and 2", Supplement 10, NUREG 0797, Docket Number 50-445 and 50-446, April 1985. 9.2 TUCCO QC Inspection Procedure, QI-QP-11.18-3, Revision 4, 

                                     " Ultrasonic Thickness Measurement".

9.3 Project Operation Travelers: CE-86-222-8902 CE-86-223-8902 CE-86-224-8902 O CE-86-225-8902 9.4 Project Operation Travelers" CE-87-1934-01-8902 CE-87-1935-01-8902 CE-87-1938-01-8902 CE-87-1939-01-8902 CE-87-1936-01-8902 CE-87-1937-01-8902 CE-87-1940-01-8902 CE-87-1941-01-8902 9.5 Stone & Webster Engineering Corp. Specific Technical Issue Report, STIR-CPRT-S-001, " Threaded Fastener Evaluation", Revision 0. 9.6 Memorandum, J. Arros to ISAP V b file, dated 10/15/87, " Third Party Review of V.b Inspection Procedures and Project Operation Travelers". 9.7 Memorandum, J. Arros to ISAP V.b file, dated 9/24/87, " Steam Generator Upper Lateral Restraints". 9.8 " Summary Report, Steam Generator Upper Lateral Support" h V Southwest Research Institute, October 1987. 

                                                                                                    )

l __ f

l Revision: 0 I Page 55 of 62 I RESULTS REPORT ISAP V.b (Cont'd) i

9.0 REFERENCES

(Cont'd) 

                        ~9.9    AFC0 Drawings:

77668-F1 Revision 3 77668-101 Revision 2 77667-303 Revision 2 9.10 Memorandum, C. Trout to J. Arros, dated 10/14/87, " Comanche Peak Steam Electric Station SGULs Beam Rework ISAP V.b I Information". 9.11 G6H Calculation SRS-167C, Set 4, (Not issued). 9.12 Memorandum, J. Arros to ISAP V.b File, dated 10/12/87, " Third Party Review of G6H Calculation SRB-167C, Set 4". i 9.13 Memorandum, J. Arros to ISAP V.b File, dated 10/14/87, ' 

                                " Comparison of Calculated Thread Engagement Requirements to-the Actual Measured Thread Engagements".

9.14 Letter, M. J. Holley to H. Levin, dated 7/18/85, " Upper ' Lateral SG Restraint Beam Connections to R/C Walls". 9.15 Memorandum, M. J. Holley to J. Arros, dated 7/28/87, 

                               ~" Evolution of Loads on the SGUL Connection Bolt Groups and the Underlying Analyses".

9.16 G&H Calculations: LIS-1090, Set 1, Revision 1 LIS-109C, Set 2, Revision 1 , LIS-109C, Set 3, Revision 0 LIS-109C, Set 4, Revision 2 LIS-509C, Set 1, Revision 0 LIS-509C, Set 3, Revision 0 LIS-510C, Set 1, Revision 0 LIS-5100, Set 2,. Revision 0 I SRB-167C, Set 1, Revision 1 l 9.17 Westinghouse Report, " Comanche Peak, Inspection and Evaluation for Steam Generator Upper Support Beam Anchor Bolts", Revision 2, December 1986. 9.18 Memorandum, J. Arros to ISAP V.b File, dated 10/15/87, " Third i Party Review of Westinghouse Report on SCUL Bolts". I

i i Revision: 0 l Page 56 of 62 S g v) RESULTS REPORT ISAP V.b (Cont'd)

9.0 REFERENCES

(Cont'd) 9.19 Nonconformance Reports (and associated Field Design Change-Change Verification Checklists): CM-87-1332 CM-87-1333 CM-87-6420 CM-87-6919 CM-87-6924 CM-87-6921 CM-87-6922 CM-87-6925 9.20 TU ElectricPurchase Order CPF-13865-S (nith Supplements) to Southwest Research Institute. 9.21 Memorandum, J. Arros to ISAP V.b File, dated 9/9/87, " Steam Generator Upper Lateral Restraint Documentation". s 9.22 Brown & Root Purchase Order No. 35-1195-14915 (with amendments), to AFC0 Steel. 9.23 Contact Log Sheet, 10/2/85, telecon between R. Melton (TERA) and R. Oates (AFCO), " Fabrication Tolerances of Steam Generator Upper Lateral Restraints / Supports". 9.24 Contact Log Sheet, 10/31/85, telecon between R. Melton (TERA) and R. Oates (AFCO), " Steam Generator Upper Lateral Restraints". l 9.25 Memorandum, J. Arros to ISAP V.b file, dated 10/24/86, "Walkdown to Determine SGUL ID Markings". 9.26 Brown & Root Receiving Inspection Report, RIR-05578, 2/27/78. 9.27 Brown & Root Receiving Inspection Report, RIR-05579, 4/1/87. l 9.28 Brown & Root Receiving Inspection Report, RIR-05838, 4/1/87. 9.29 TUSI Letter, from B. J. Murray to R. Tolson, dated 4/18/78, 

                                 " Comanche Peak Steam Electric, 1981-83, 2300 MW Installation, NCR M-906 and M-907".

9.30 Memorandum, J. Arros to ISAP V.b file, "11/11/86 Meeting with ( s) Project Personnel". i l t 

                                                                                         . _ _ _ _ _ _ . _ _ _   _ _ _ _ _ _ _ _ - - - _ J

Revision: O Page 57 of 62 

 /^)
 \

RESULTS REPORT ISAP V.b (Cont'd)

9.0 REFERENCES

(Cont'd) 9.31 TUCCO Letter, from W. A. Kattness to M. D. Warner, dated 9/16/84, " Installation Inspections of the Steam Generator Upper Lateral Supports". 9.32 QA/QC Review Team ISAP VII.c Results Report, " Construction Reinspection / Documentation Review Plan", Revision 1. 9.33 " Manual of Steel Construction", AISC, 7th Edition, 1970. 9.34 CPSES Concrete Pour Card, No. 101-4832-004. 9.35 CPSES Concrete Pour Card, No. 101-7860-009. 9.36 Memorandum, J. Arros to ISAP V.b File, dated 9/15/87, " Review of SWEC, Appendix I, STIR-CPRT- S-001, Revision 0 (7/15/87)". pO l

Revision: 0 Page 58 of 62 (} RESULTS REPORT ISAP V.b (Cont'd) Table 1 Summary of Related DIRs DIR # SUBJECT CLASSIFICATION D-0003 Anchorage of SGUL Embedments Originated in l DSAP VIII D-0017 Gibbs & Hill / Westinghouse Interface Transferred to DSAP VIII D-2152 SG Restraints - DCAs & Lamellar Tearing Observation D-2476 SGUL Thread Damage and Grinding of Solt Unclassified Threads Deviation D-2477 Lack of Thread Engagement Requirements in Unclassified Design Deviation D-2478 SGUL Shim Pack Thickness Unclassified Deviation D-2479 Transfer of Vendor Information to Design Unclassified Deviation D-2480 Procedural Control of Match-Mated Unclassified Installations Deviation D-2481 Construction Failures during SGUL Unclassified Installation Deviation D-2482 Lack of QC Inspection Requirements for Unclassified Thread Engagement Deviation D-2483 QC Identification of SGUL Installation Unclassified Problems Deviation D-2484 Lack of SGUL Thread Engagement Unclassified Calculations Deviation D-2485 Lack of QC Documentation Unclassified Deviation E-0990 Steam Generator Support Anchor Bolt Unclassified Shortening Deviation O1

Revision: 0 Page 59 of 62 7-( j RESULTS REPORT 

        'v ISAP V.b (Cont'd)

Table 1 (Cont'd) DIR # SUBJECT CLASSIFICATION E-1055 Cutting of SGUL Bolts, due to Concrete in Unsubstantiated Embedment Hole E-1056 Unauthorized Cutting of SGUL Bolts Unclassified Deviation E-1208 SGUL Connection Design Merged with D-2484 l E-1224 SGUL Bolts Unclassified Deviation E-1211 Steam General Upper Lateral Restraints Merged with D-0003 E-1209 Steam General Lateral Support Embedments Merged with D-0003 i 2 i _______ _ _ _ _ _ _ _ _ _ _ _ _ l

7 1i ! Revision: 0 Page 60 of 62 l RESULTS REPORT l j ISAP V,b l (Cont'd) Figure 1 t i 

                                       ,,,y.i' *.T5 5 [p.'.'f.f d?.f. ; ;,*.[ ?.ij *2.*i.'.,' I;*;.*'.'s,?.i il, . .

t*.. 

                                                                                         *. .6.'
                                                                                                                                       *4.* '.j                                                               l 1
                          .* * *                                                         ,.f';
  • y . ..
                        .                                                      '.g...
                                                                                                                                              . ' $, g .
                   .,,.                                                        :;. .'.?.                                                                   .y. .
              . .e.
                                                \                    l           .;. ..*,

i 1  ;. .

i. , ..;. + ;.
            ,<                                                                 ,*      *g*

y ,,.. 

                                                                                                                                                                         *1.-'*
          ,( <i                                                                t. '< .-

l,*.' j' * 

                                                                               ,j              ,

E' .;. 

          '.s                           .

m.g. ....; - - - 37.', E.SU'.!iO9.}.M9h?' b... '* ..,.; . gp". 7s. 

                                                                                                             ' $..wm *).D (Q.UMT.MMpf
                                                                                                                                                                                 ;(          '

500L  :, Reetraint Beam 0 . Rin, easm  : 

                                                                                                                                                                                 .';:       m Y

eam oenerusor l { actor & Containment { Reactor e. y Containment Bido. 1 I f'*.!)* .$b . b *$' s$.*.' k.*gA. * 

                                                                                                                     .h,.h.,*a.g*)a.'.*.*5'?  O' ,'J** .' ',/ EN
                                                                                                                                                                             . .E* I. [.
                                                  .,f-,s}.,j
                                                        , , ,g, .e,j.
                                                                , ,. , ,. gt ..,,.
   ..s,..e'...                                                                       f 3,, .,;*.'fg,*,.,* *, . g ......
  .*                                                                                        , . ;L....*....,e
                                                                                                  . , e J 4,.., ..*.1.-

r..,.,=, 

 '.~
 >:.        ..                                                             f.'.d     N                                                                                           . . , ..
                                                                                      .s. ,                                                                                                                    ,
  *..                                                                        .e.,    ' .* .
  • l
                                                                                                                                                                     .! , .*ye,.
    *d' , .                                                                *'                                                                                                                                  \
                                                                                                                                                                                   *l.  .

1 

              ..                              t                    l       +'
                                                                           .e         !
                                                                                       .              \                  \
                                                                                                                                                             ..u.
                   ..                                                      +..
                                                                                                                                                       .      .o
                . * ..,. . < ..v. 6
                                                                                                                                                   . :,s.   . . .. .
                                                                           '. g*
                                                                                                                                                                                                              .i
                           ! .,                                            ,1 , * . . * * ,                                            . . - *:

l 

                                                                                                                         . .. . .. . 4 s. .
                             ' ? ;-                                         .,1 ' .1
                           .g..,
                                          ,                                                                                                                                                                    I
                                    .e                                     ?- t                                        .       *
                                         ' h')hh 'h',f*h*.' *                .
                                                                                            *'. ).*.h.', h            .
  • U I

1 i

Revision: O Page 61 of 62 RESULTS REPORT ISAP V.b (Cont'd) Figure 2 V Dk < 

                                                             .. by. i.

S( '. ' M e (See F>us 3) Y O -- - End Piste m Pack Embedmont h P '%k. O , l 1

Revision: 0 Page 62 of 62 r RESULTS REPORT ISAP V.b (Cont'd) Figure 3 t o 1/2* , 

                                                                             ==        e.                                                         1 O O O O O                 U
r- ..

es , to r 1 B 4 O$ $0 q 3*' O!s $0 s l-O!! k,On- 2 seis og k e . . ..  : O -,- 

                                                                                                                                                   )
                                                                                                                                                    \

Section A A I I l 1 Ca 4 1/4* Min. I E c! Ca Ce l l Ca _ Section B-B 

                                                                                                        . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _   -_-}}
Retrieved from "https://kanterella.com/w/index.php?title=ML20236X830&oldid=3170470"