ML20235S177
| ML20235S177 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 09/23/1987 |
| From: | Beck J TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC) |
| To: | |
| Shared Package | |
| ML20235S175 | List: |
| References | |
| NUDOCS 8710080356 | |
| Download: ML20235S177 (43) | |
Text
- _ _ - _ _ - .
Vi u . .,
p .-
l t .
COMANCHE PEAK RESPONSE TEAM RESULTS REPORT ISAP: VI.b
Title:
Polar Crane Shimming REVISION 1 i
- O
/M Isdue Coordinator f//s//7 Dat( /
- m\
luevieV Team Leadir'V okle, Date'
& N. 14_ $f2.3/V1 JohgW. Beck,ChairmanCPRT-SRT Date l-8710080356 8k45 PDR ADOCK PDR
^
O
- Revisien: 1 4
Pega 1 of 42
,_ RESULTS REPORT t i N/ - . ISAP VI,b Polar Crane Shimming
1.0 DESCRIPTION
OF ISSUE IDENTIFIED BY NRC Issue VI.b was identified in SSER 8 (Reference 9.1, pages K-181 and K-182) as follows:
The TRT investigated the installation of the polar crane rail support system by visual inspection, review of associated documentation, and discussions with TUEC representatives and j their contractors. Region IV Inspection Report 50-445/84-08; 50-446/84-04, and Notice of Violation, dated July 26, 1984, documented that gaps on the Unit 1 polar crane bracket seismic connections exceeded design requirements. In Texas Utilities Generating Company responses of August 23, 1984, and September 7, 1984, the gaps were attributed to crane and bolting self-adjustment resulting from crane operation. A site design change (DCA-9872, Revision 4, dated August 24, 1984) was issued to document the acceptability of gaps in excess of 1/16 inch which were identified in the above NRC inspection report.
y During further investigation of the allegation that shims for the rail support system nad been altered during installation, the TRT observed gaps which may have been excessive between the crane girder and the girder support bracket. Detailed specifications addressing the gap tolerances in the girder seat connection did not exist; however, Gibbs and Hill letter GHF-2207*, dated November 28, 1977 had stated that the " seated connections will not require shimming since the area in bearing is at least the width of the bottom flange of the crane girder." Contrary to this Gibbs and Hill assumption, the TRT observed nine girders with gaps which extended under l
the bottom flange that reduced the bearing surface to less than the 20 inch flange width stated in the letter. The TRT also observed conditions which indicate that the crane rail may still be moving in a circumferential direction, that three rail-to-rail ground wires were broken, that two shims had partially worked out from under the rail, and that two Cadwelds were broken.
2.0 ACTION IDENTIFIED BY NRC i
The actions to be taken regarding Issue VI.b were identified in SSER 8 (Reference 9.1, page K-182) as follows:
10
() s,
- GHF-2207 transmitted a copy of letter DALM-209. The statement -
" seated connections . . . crane girder." - appears in DALM-209, not GHF-2207.
l ' Revision: 1 L ,
Pego 2 of 42 1.
RESULTS REPORT
(
(m / . -
ISAP VI b (Cont'd) l 2.0 ACTION IDENTIFIED BY NRC (Cont'd)
Accordingly, TUEC shall:
- 1. Inspect the polar crane rail girder seat connections for the presence of gaps which reduce the bearing surface to less than the width of the bottom flange and perform an analysis which will determine whether the existing gaps are acceptable or require corrective action.
- 2. Determine if additional rail movement is occurring and, if so, provide an evaluation of safety significance and the need for corrective action.
- 3. Perform a general inspection of the polar crane rail and the rail support system, correct identified deficiencies of safety significance, and provide an assessment of the adequacy of the existing maintenance and surveillance programs.
3.0 BACKGROUND
The Unit 1 polar crane is located in the containment structure and operated on a circular runway system. The primary function of the crane is to lift loads during refueling and maintenance operations, but it was also used to install NSSS components and to assist in other construction activities. Its rated capacity for the operations phase of the plant is 175 tons; however, during the construction phase it was rated for 475 tons.
The polar crane and runway systems have been classified as Seismic Category I, single-failure proof, with the capability to retain control of a lif ted load and maintain structural integrity during and after an Operating Basis Earthquake (OBE), and to maintain structural integrity without lifted load during and after a Safe Shutdown Earthquake (SSE). During normal operation of the plant, the crane is set in a pre-established parked position.
The major components of the crane are the hook, hoist, trolley, two bridge girders, four bridge trucks and sixteen wheels. The wheels are double-flanged; the flanges serve as mechanical guides to prevent the crane from being derailed when operating in a normal manner or when subjected to a maximum loading combination, including an SSE.
O
o -
R3 vision: 1 Peg 2 3 of 42 RESULTS REPORT f~3
(/ - _
ISAP VI.b (Cont'd)
3.0 BACKGROUND
(Cont'd)
The runway system consists of 14 segments of Bethlehem 171 lb/yd crane rail, curved to a 65 foot radius, and supported by 28 crane 1 runway girders (hereafter referred to as " girders"), which in turn seat onto crane runway support brackets (hereafter referred to as
" support brackets"). The support brackets are anchored to the )
concrete containment wall and welded to the liner of :he containment structure (see Figures 1 & 2). Seismic brackets are also anchored to the containment wall and welded to the liner, and provide radial and longitudinal restraint at the midpoint of the ,
j girder. The rails were not originally spliced together, but a modification program is underway to add splice bars to limit the amount of gap between rail ends. The rail was originally attached to the girder by the rail clips shown in Figure 3. The rail clips j were designed to allow limited circumferential rail movement while providing restraint of vertical and radial movement. ];
The girders have four important components (Figure 2): 1
- 1. Girder radial restraint attachments (hereafter referred
[~'}
\) to as " radial attachments") mate with the upper support brackets and permit girder motion in the vertical direction and longitudinal direction of the girder, but restrain movement in the radial direction.
- 2. Girder seismic restraint attachment (hereafter referred to as " seismic attachment") that mates with a seismic bracket and permits movement in the vertical direction, but restrains movement in both the longitudinal and the radial directions.
- 3. Sole plates that provide an attachment for the girder to the support bracket. The sole plates permit girder motion in the longitudinal direction, but restrain movement in the vertical and radial directions.
- 4. Uplift flange that was intended to restrict any seismic uplift motion of the crane by engaging one of four seismic lugs attached to the ends of the cranr. bridge girders. (SWEC has determined that there will not be uplift; therefore, the flanges will not te relied upon.)
Support bracket installation was performed by Chicago Bridge and Iron (CB&I) as part of the containment liner contract scope. The girders were set in place after the support brackets had been O temporarily welded on location. This was to provide permanent
. Revision
- 1 Pega 4 of 42 s RESULTS REPORT I )
~
ISAP VI.b (Cont'd)
3.0 BACKGROUND
(Cont'd) alignment during final welding of the support brackets to the liner. Shimming of the girder seat connection (the attachment between the girder sole plate and the support bracket) was performed concurrently with these activities.
During girder installation, gaps were encountered between mating surfaces at all connections (girder seat, girder to upper support bracket, and girder to seismic bracket). An engineering assessment was conducted to determine the design 11gnificance.
After field inspection, Gibbs & Hill (G&B) indicated that girder seat connections would not require shimming since the area in bearing is at least the width of the bottom flange of the girder.
However, G&H required that upper support bracket connections and the seismic bracket crenections be shimmed in accordance with G68 sketch SK-82032. This information was forwarded to G&H-Site in letter number DALM-209 (Reference 9.2) and later issued as Design Change Authorization (DCA)-9872. An allegation (Reference 9.3)
('~T concerning installation practice of these shims brought about a
\_ / reinstallation and reinspection of the shims.
It was later determined that the upper support bracket and the seismic bracket connection gaps had increased. The effect was assessed and DCA-9872 was revised to accept gapc in excess of 1/16", which eliminated the need for further shimming. However, during the TRT inspection of the runway support system it was observed that nine girders had gaps in the girder seat connections that extended under the bottom flange. These gaps are addressed as part of this action plan.
As-built measurements of the rail were provided as part of the initial CB&I documentation. These as-builts included diameter and elevation of the rail. Friction clips were added to the rails after installation per DE/CD-S-1855 and DCA-6437 in an attempt to prevent circumferential movement of the rail. As non-nuclear safety-related components, such items receive no inspection of bolt tightening. After installation of the friction clips, rail movement along the circumference continued to be observed. Rail creep restraints were then designed and installed per DCA-15337.
These restraints were found to be ineffective. On closer inspection it was found that the friction bolts were loose. These bolts were then torqued under engineering direction. Precise rail locations were not documented so base line data was not available, y-~ for making precise determination of subsequent rail movement.
i
- Revision: 1 Page 5 of 42 RESULTS REPORT
'l '
\ -
ISAP VI.b (Cont'd) 4.0 CPRT ACTION PLAN The objectives of this action plan were to:
Determine if the gaps in the girder seat connection are acceptable and provide corrective action if necessary.
Determine if circumferential crane rail movement is continuing and, if so, assess its significance and the need for corrective action.
Provide additional. assurance, through a general reinspection of the runway system, that any deficiencies have been identified and corrected.
Assure that the polar crane maintenance and surveillance programs include those aspects of the rail and runway system that should be periodically reinspected.
4.1 Scope and Methodology
() The objectives of this action plan were addressed by the following tasks:
4.1.1 A' review of the history of the runway system fabrication' construction, and performance during operation in the construction stage was conducted. The design requirements were identified and categorized as operational performance requirements (e.g., rail alignment) or structural adequacy requirements (e.g.,
structural integrity).
4.1.2 A screening visual inspection of all girdar seat connections, followed by measurement and mapping of the actual bearing area of the worst case gaps, was performed by SWEC to determine if the gaps present reduced the bearing surface to less than the width of the bottom flange of the girder. A third-party review of the findings assessed the safety significance and determined whether the design requirements had been met. Corrective action to increase the bearing surface was to be prescribed, if necessary.
O
Revision: ;l Page 6 of 42 RESULTS REPORT (s / -
ISAP VI.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd) 4.1.3 A measurement program to determine the extent of rail movement was developed and executed by the Project and the third party in order to provide data to assist in the determination of the cause of the rail movement.
Modifications in the form of rail-to-rail splice bars were initiated as a result of these programs. G6H performed an analysis and provided initial design details for the splice bars. SWEC subsequently completed an analysis and a final design of the splice bar installation. The analysis and design were reviewed by the third party.
4.1.4 Criteria for a general inspection of the rail and runway system were established by the third party. The general inspection was performed by the Project and overviewed by the third party.
4.1.5 The third party reviewed the existing maintenance and f
surveillance program for the polar crane and runway
( system and recommended changes to the program.
4.2 Participants Roles and Responsibilities The following organizations and personnel participated in this effort:
4.2.1 TUGC0 Nuclear Engineering (TNE) - Civil / Structural Discipline 4.2.1.1 Scope Developed physical measurement plan for girder seat connection bearing.
Identified design performance requirements and construction operational history and performance of the polar crane.
I l
I
.. 1 Revision: 1 Page 7 of 42 l
RESULTS REPORT ,
/ {
ISAP VI.b .I (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd) 1 Evaluated and improved the {
surveillance program for operations. I l
Developed a plan for and performed a l general inspection of the runway j system.
4.2.1.2 Personnel Mr. C. R. Hooton THE Civil / Structural i Discipline Supervisor j Mr. M. L. Osterman Civil Engineer I
Mr. S. G. McBee Civil Engineer 4.2.2 Gibbs & Hill 4.2.2.1 Scope Performed initial analysis to 1 resolve the girder seat concerns and {
to provide design of proposed 4 modifications for correcting rail l movement problem.
4.2.2.2 Personnel 4
Mr. E. L. Bezkor Job Engineer
)
i Mr. A. M. Kenkre Squad Leader 4.2.3 Stone & Webster Engineering Corp. (SWEC)
(Lead Contractor responsibility for this task was transferred by TU Electric from G6H to SWEC on October 13, 1986. After that date, SWEC participated in the execution of this action plan as described below.)
4.2.3.1 Scope Performed a revised polar crane load analysis.
O
r-4
'- Revision: 1 Page 8 of 42 RESULTS REPORT-
) -
ISAP VI,b (Cont'd) 4.0- CPRT ACTION PLAN (Cont'd)
Analyzed and designed modifications:
required to correct rail movement problems, i !
Analyzed the original rail clips and, designed new, higher capacity rail clips.
Performed analysis to assure the adequacy of the girder seat connections.
4.2.3.2 Personnel Mr. W. N. Kennedy Engineering Support Group Assistant Manager '
Mr. J. A. Lebruto' Responsible Engineer I\ Mr. P. H. Titus Senior Principle Mechanical Engineer-4.2.4 TUGC0 Quality Assurance 4.2.4.1 Scope Inspected rail modifications, after installation.
4.2.4.2 Personnel Mr. P. E. Halstead Manager, Quality Control 4.2.5 Third-Party Activities 4.2.5.1 Scope Developed, executed, and evaluated test data for rail movement.
Investigated cause of rail movement and recommended modifications to correct the problem.
O -
Overviewad the general inspection of the runway system.
' Revision: 1
~
Page 9 of 42 1
,- RESULTS REPORT
() -
ISAP VI.b I (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd)
Evaluated significance of gaps in the girder seat connections.
Recommended changes to maintenance and surveillance programs.
Reviewed the design and evaluated resolutions.
Prepared Results Report.
4.2.5.2 Personnel Mr. H. A. Levin TENERA, Civil / Structural Review Team Leader j 1
Mr. J. C. Miller TENERA, TRT Issues Manager
)
r'N
('~ ) Mr. R. W. White TENERA, Structural 1 Engineering Consultant i
Prof. M. Holley TENERA Consultant, Hansel, Holley & Biggs l
Dr. J. K. Arros TENERA, Issue Coordinator 4.3 Personnel Qualification Requirements Third-party participants in the implementation of this Action Plan meet the personnel qualification and objectivity requirements of the CPRT Program Plan and its implementing procedures.
Where tests or inspections required 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 Persennel at Nuclear Power Plants". CPRT third-party inspectors have been certified to the requirements of the third-party employer's Quality Assurance Program, and trained to the applicable inspection procedures.
l
[]
Revision: 1 Page 10 of 42 RESULTS REPORT ISAP VI.b (Cont'd) 4.0 CPRT ACTION PLAN (Cont'd)
}
Other participants were qualified to the requirements of the 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 nuality", of the CPRT Program Plan.
4.4 Procedures (
\
The screening inspection of the girder seat connections, to identify and measure two connections representative of the {
1 minimum bearing' area, was performed in accordance with the procedure described in Attachment A of the SWEC calculation 16345-EM(S)-008-CZC (Reference 9.4).
The rail movement measurement program is described in the i Southwest Research Institure report (Reference 9.5); the test procedure is described in Appendix H of the report. j I
.V The polar crane general inspection was conducted, and the (
results reported, in accordance with Operation Traveler CE-85-2876-8902 (Reference 9.6).
4.5 Standard / Acceptance Criteria The applicable standards / acceptance criteria for evaluation of the runway system are identified in the AISC " Manual of Steel Construction," (Reference 9.7) and " Specifications of Electric ;
and Travelling Overhead Cranes," CMAA Specification #70, (Reference 9.8). j 4.6 Decision Criteria The decision criteria for the inspection and evaluation of the girder seat connection gaps and the engineering significance j of the gaps are that, regardless of the pattern of the contact bearing surface (full or partial), the runway girder and all ,
l connecting hardware shall satisfy the stress and deflection j criteria established in the governing specifications. If '
these criteria are not met, design modifications will be !
undertaken to correct any deficiencies.
l
Revision: 1 Pego 11 of 42 RESULTS REPORT l
V' -
ISAP VI.b (Cont'd) i l
5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS The principle concerns in this issue are the existence of gaps at girder connections and the circumferential movement of the rail.
In the course of its investigations of both issues, the third party '
determined that the G&H design calculations for the runway system were incomplete in some areas. After the transfer of 1 civil / structural design responsibility from G&H, SWEC reviewed the 1 G&H analyses and, where appropriate, developed new design calculations.
The results of third-party investigations relating to the concerns of this issue are covered as follows. Section 5.1 describes the design, installation, operation, and inspection history of the runway system; Section 5.2, the evaluation of the girder seat connection; Section 5.3, the significance of the circumferential 1 movement of the rails; Section 5.4, the general inspection of the runway system; and Section 5.5, a review of the maintenance and surveillance program requirements related to the polar crane and runway support system. Section 5.6 describes an out-of-scope
()
r"%
observation. A summary of DIRs is provided in Section 5.7, and safety significance evaluation is provided in Section 5.8. Section 5.9 addresses the root cause and generic implications.
Design modifications that resulted from the investigations of Unit I will also be performed on Unit 2.
5.1 Review of Design Requirements and History of the Runway System l The review was performed by the third party with assistance !
from the Project and resulted in documentation of the design l and installation requirements for the runway system, load testing of the crane, and the installation and inspection history related to the shim gap and rail motion problems.
The requirements for various parameters of the crane are listed in Table 1. Reference 9.9 presents a history of the crane and runway system from initial fabrication and construction through activities occurring in 1986, a summary of which is contained in Section 5.1.2 below.
5.1.1 Runway Girder Design Prior to the TRT Investigation The engineering design of the runway system prior to the TRT investigation is documented in G&H Calculation SRB-109C, Set 3, Revision 0 through 3. The interface
(,,
\
wheel loads for the various load combinations, including seismic, used in the G6H analyses were provided by KRANCO, the crane manufacturer.
l l
l
- 1
' Revision: 1 Page 12 of 42 l
RESULTS REPORT
' ]
ISAP VI,b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) )
Significant aspects of the runway system' design that are not addressed by SRB-109C, Set 3, Revision 0 through 3 are:
Rail stresses and rail sizing criteria.
Design of original rail clips and welds, Design of the sole plate, sole plate-to-girder welds, and sole plate bolts, Calculation of girder connection contact bearing stresses and calculations and criteria for the required bearing areas, and Original member sizing calculations for plate thicknesses, fabrication weld sizes, and the location of both external and internal stiffeners.
These discrepancies were identified in DIRs for resolution by SWEC. In addition, G&H completed analyses of the structural members and connection details in calculation SRB-109C, Set 3, Revision 4 through 10. These analyses will be reviewed by SWEC as part of the design validation activity of the Corrective Action Program.
5.1.2 Installation of the Polar Crane The installation of the support brackets and girders and the fitup welding and inspection of the upper brackets and seismic bracket connections were performed by CB&I beginning in July 1977. Brown & Root (B&R) provided the layout and elevation control of the runway {
support brackets.
Elevations of all support brackets within one quadrant of the containment building were determined and the support brackets were attached with temporary welds to the liner shell to allow a proper fit-up with the simultaneous installation of the girders. The first O
w Rsvision: 1 Pags 13 of 42 RESULTS REPORT i j -
ISAP VI.b j (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) girder was placed on the highest installed support bracket, and brought to elevation at the girder seat connection with shims. The remaining girders were installed and their seats were shimmed to obtain an elevation equal, or within allowable tolerances, to that of the first installed girder.
The installation of girders and support brackets was completed in August 1977. The rail and its attachments were in place by February 1978, and final rail positioning was completed in March 1978.
Girder Seat and Upper Bracket Connection Gaps During the insta11atien of the girders and the shimming of the girder seats to maintain elevation, G&H-Site noted that fu11' contact area was not being maintained due to variations la alignment and construction tolerances. As a result of discussions between B&R and O- G&H, it was agreed that machined shims would be used to correct the lack of contact. (Reference 9.10).
i In November 1977, prior to the installation of the crane, as-built documentation of the gaps for all girder seat connections and upper bracket and seismic bracket connections were provided by B&R to TU Electric (Reference 9.11). After a review of thin :
documentation, as well as an on-site visit, G&H stated '
(Reference 9.2) that the girder seat connections did not require any further shimming since the area in bearing was at least the width of the bottom flange of the girder. However, G&H did require that shims be installed at the two radial restraint attachments and at the seismic restraint attachments of each girder, but stated that residual local gaps of 1/16 inch are acceptable (Figure 2). The shims were installed by DCA 9872 Revision 0. i An NRC inspection of the polar crane system was conducted between April and September, 1982. In June 1982, the NRC Senior Resident Inspector for Construction (SRIC) determined that no inspections of the fabrication or installation of certain shims were performed. In addition, the SRIC was informed that an O allegation had been made by a former construction
_ _ ___ _ __ _ -- - -- _ ----- ~ - ------ - --- - ----- -- ---- ---------~-- ~-~ - ~ ~ - ~
~
Revision: 1 Page 14 of 42 RESULTS REPORT ISAP VI.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) worker that shims for the runway system "had been altered during installation in some unacceptable manner" (Reference 9.3). In July 1982, a Notice of Violation was issued to TUGC0 (Reference 9.12).
Shortly after the issuance of the Notice of Violation, TUGC0 QC issued Non-conformance Report (NCR) M-82-894 concerning the upper brackets (Reference 9.13).
In August 1982, DCA-9872 was revised to clarify 5 shimmir g and bolting requirements, and an operations traveler CE-82-370-8104 was issued by TUGC0 to implement the QC reinspection of the upper bracket shims. As a result of this reinspection some reshimming was performed by craft personnel and inspected by QC. The NCR was closed in January 1983.
In February 1983, DCA-9872 was revised to require that any upper bracket connections with gaps in excess of 1/16 inch be reshimmed. During an NRC inspection from November 1983 to March 1984, gaps in excess of 1/16 inch were observed in the upper bracket connections and as a result, the NRC issued a Notice of Violation (Reference 9.14).
In January 1984, TUGC0 notified G&B of the concerns expressed in the NRC inspection reports. G&H's evaluation in February 1984 noted that gaps about the periphery of the shims are of no concern as long as there is positive contact at the girder to allow for transfer of seismic loads directly into the containment wall (Reference 9.15). SWEC confirmed the adequacy of the upper bracket shimming by issuing Revision 5 of DCA 9872.
In March 1984, the effects of a new set of as-built data on the upper bracket connection were assessed by G&H and reported to the site in GTT-10372 (Reference 9.16), along with a recommendation that a program of periodic gap measurements be established in order to monitor the condition. DCA-9872 Revision 4 was revised in August 1984 to accept gaps in excess of 1/16 inch.
Also, in August 1984, G&H suggested to TUGC0 that nine of the upper bracket connections could be reshimmed to avoid future monitoring (Reference 9.17). No action was taken by TUGC0 on the suggestion, and no further O- reshimming of the upper bracket connection was
Revision: 1 Paga 15-of 42 RESULTS REPORT f).
'- , ISAP VI.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) performed; however, inspection of upper bracket connections was later included in the plant mechanical inspection procedures-(Reference 9.18).
The TRT, during an inspection of the runway system in August 1984, found that nine of the girder seat connections had gaps that extended into the projection of the bottom flange of the girder. The gaps exceeded the acceptance criteria originally established by G6H in November 1977 (Reference 9.2).
Rail Motion In October 1979, site personnel votified G6H NY that the rails had moved and that a 3-1/4 inch gap existed between two rails (Reference 9.19). G&H responded that corrective action would be necessary and proposed that four friction clips be added near the center of each rail section (Reference 9.20). These clips were installed in accordance with DCA-6437, Revision 0, in January 1980. Figure 3 shows a typical friction clip.
In June 1982 an allegation was made to the SRIC that the rail moves during crane operation (Reference 9.3).
In discussions with TUGCO, the SRIC was advised that the addition af the friction clips, and a correction at the crane truck wheels to resolve binding between the wheels and the track, would correct the problem.
In December 1982, additional rail movement was identified. In response, TUGC0 Engineering issued DCA-15337, Revision 8. This DCA resulted in the installation of a 1-inch-diameter pin through the rails, held by brackets to the top of the runway girder (Figure 3. Rail Creep Restraint). However, observations of bent rail creep restraint pins made during its Inspection in August 1984 led the TRT to believe that circumferential rail movement was still occurring (Reference 9.21).
O 9
Revision: 1 Pegs 16 of 42
,x RESULTS REPORT
! t
\
/ '
IS AP VI.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) 5.1.3 Operational History of the Polar Crane After installation of the runway system was completed, the crane was installed in January 1978 and tested in April 1978 (Reference 9.22). The crane lifted a total of 499 tons in the load test and was rated for 475 tons for the construction phase. In August 1982 the crane was down-rated to 175 tons for the operational phase.
]
Wheel-to-Rail Interface After the runway system was completed and the polar crane was installed, it was observed that the truck cheel alignment of the crane with respect to the rail was causing a binding action between the wheels and the rail. To resolve the problem, KRANCO modified the bearing cap orientation of the crane truck wheels to improve the alignment with the rail configuration. New
/N bearing caps were installed and accepted in December
(--) 1979. Following this modification, the crane operators contacted by TUGC0 reported that the crane exhibited only minor vibration and rail noise during rotation.
Third Party Observations In February 1986, the third party confirmed earlier project observations that the wheels and rails exhibit only minor levels of noise and vibrations during operation. Overall, the performance was reported to be generally smooth except when the operator performed a simulated emergency stop at the request of the observers. This required bypassing the normal five j speed shift down to neutral (brake), bringing the crane ;
to an abrupt stop, causing increased crane and rail i vibration and rocking back and forth after stopping, as was expected. The normal mode of downshifting to a full stop results in essentially no vibration. ,
I Starting the crane from rest was characterized by a sudden initial jerk, but with only minor vibration.
Operators, certified to Maintenance Department .
Administrative Procedure KDA-308, " Crane and Electric l Hoist Certification Program", use the normal mode of r' s\ downshifting to a full stop.
i v
l
. Revision: l' ]
Page 17 of 42 l-
~
I RESULTS REPORT
>(N y -
ISAP VI.b (Cont'd) l 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd)l In May 1986 the head of the Unit I reactor was removed, lifted, and then positioned onto a temporary storage platform. This event was witnessed by the third party.
The digital readout on the crane indicated that approximately 157 tons had been lifted and moved.
j During this time the third party closely observed the i behavior of the shimmed runway girder seats at a readily accessible runway girder support bracket. (
There was no visible indication of gap closure on this j or adjacent seats as the loaded crane passed over the seat. {
In August 1986, third party and Project personnel rode the polar crane with the intention of observing whether or not any shim gaps could be observed as opening or closing. No visible shim gap movement was observed.
5.1.4 Inspection History of the Runway System j
h The Comanche Peak Steam Electric Station Maintenance Manual HMI-317, Revision 1, (Reference 9.23) addressed only the " periodic inspections of the Containment Polar Crane." There were no requirements for the i inspection of the girders and support brackets, and I thus no periodic inspection was scheduled or performed '
for the girders, rail, and support bracket.
5.2 Girder Seat Connection The circumstances leading to the investigation of the girder seat connection have been described in earlier sections of this report. In order to confirm the adequacy of the existing connections, SWEC performed the investigation described below.
5.2.1 investigation Determination of Bearing Contact At the request of SWEC, a craft team performed a visual screening of the 28 girder seat connections, identified four connections representative of the minimum bearing contact, and performed a detailed inspection to l determine the actual bearing areas. These data were then used in the analysis of actual bearing contact locations, and to provide an insight into the validity
(
of assumptions used in developing the analysis discussed below.-
' R2 vision: 1 Pag 2 18 of 42 RESULTS REPORT (m / -
ISAP VI.b (Cont'd)
I i
5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cent'd)'
Determination of Seismic Loads An initial step in the investigation of the adequacy of the connection was to review the wheel loads provided by KRANCO and used by G&H in the initial design calculations. For this purpose SWEC performed a l
reanalysis of the Polar Crane seismic loads. .j I
The SVEC Polar Crane Seismic Reanalysis calculation
{
(Reference 9.24) utilizes a non-linear time history 1 analysis of the system consisting of the lifted weight, the cable, the trolley, the two crane girders, the ;
trucks at each end of the girders, and the wheels within each truck. The three-dimensional model utilizes nonlinear elements to model the cable behavior !
(i.e., the slack in the compressive direction) and the gaps between the rail head and the inner and outer flanges of the wheels, as well as the radial sliding of g the wheels across the gap on the rail head, with the t associated friction resistance.
s During the design of the plant, G&B generated i in-structure response spectra corresponding to different soil conditions and different assumptions of I the concrete response (cracked and uncracked). In the I Seismic Reanalysis calculation SWEC determined natural frequencies of the crane with and without the lifted load for different crane trolley and hook positions.
Based on the information about the natural frequencies. {
the in-structure response spectra that would yield the !
highest response was selected to be used for the crane l
evaluation. Time histories corresponding to these spectra were input to the nonlinear model of the crane.
This analysis predicted lower wheel loads that would be transmitted to the rails, the rail clips, the girder, and the support bracket than originally provided by KRANCO. This analysis also demonstrated that uplift of the wheels off the rail would not occur as a result of seismic excitation, as had been predicted in earlier KRANCO analyses. Reduction of the calculated vertical wheel loads and elimination of the uplift case can be explained by the fact that, in the earlier analyses by KRANCO and G&H, very few points were used to define the
(~'}
\s /
response spectra. Therefore, for large portions of the frequency range, the spectra used in the analysis
t R:visien: .1
- 4. Pag 3 19 of 42
) 1 ISAP'VI.b (Cont'd)
-5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd)
'"* . significantly exceeded the actual spectra defined by G6H.
l Reduction in the calculated horizontal wheel loads can be explained by the fact'that in the KRANCO analyseu the wheels were horizontally restrained at the rail location, resulting in a pinching effect, with the associated large horizontal loads to'the wheels. The SWEC analyses incorporate the gap elements and model the freedom of a wheel to move with respect to the rail head as far as the distance between the wheel flanges-permits.- This is a more realistic modeling assumption.
5.2.2 conclusions on Bearing Area and Contact The main implication of the gaps between the girder and the support bracket seat was the need to evaluate the bolts at the sole plate and at the upper bracket connection for the additional loads they might be
/ subjected to due to the change in the force configuration due to the gaps. The extent of the bearing area between the girder and the support bracket seat is not viewed as a concern by the third party. If the bearing area at a seat without applied loads from the crane is small, and local yielding at the contact surfaces should occur when loads are applied, the area in bearing will increase as the load is increased. No failure of the contact surfaces or the associated structures could be expected as a result of potential i local yielding at the contact surfaces.
The SWEC calculation concerning the girder seat connection (Reference 9.4) evaluates both the sole plate bolts and the bolts at the upper brackets, by assuming the worst case pattern of contact area that could be postulated. In this case, contact between the girder and the support bracket seat is postulated only at the outer edge of the girder sole plate, resulting ,
in maximum censile loads at the upper bracket' bolts and '
maximum shear loads at the sole plate bolts. Based on these analyses, all bolt stresses are within their allowable values when the rail is subjected to the loads obtained from the SWEC Seismic Reanalysis. These enveloping analyses qualify the girder seat connection bolts regardless of the actual pattern or location of contact areas at the mating surfaces.
Revision: 1
. Pass 20 of 42 4
RESULTS REPORT (Q_,j -
ISAP VI.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd)
Thus, while the investigation of the girder seat connection confirmed the TRT-observed gaps under the bottom flange of the girder, the subsequent evaluation verified that these gaps are not significant, and that no corrective action is necessary.
5.3 circumferential Movement of the Crane Rails 5.3.1 Investigations of Rail Motion In order to determine the reason for rail motion and a practical solution to the problem, the independent investigations described below were performed.
Project Investigations In February 1985, representatives from TUGCO, G&H and the third party performed a walkdown of the Unit 1 {!
,_. polar crane that included a visual inspection of the (A rail during the operation of the crane without load.
From this walkdown a report was prepared by G6H ;
(Reference 9.25) detailing the operating characteristics of the crane, suspected causes of the operating characteristics, and proposed improvements to eliminate or mitigate the most undesirable operating characteristics.
The report by G&H documented the following:
- 1) several instances of radial displacement between adjacent rail sections,
- 2) rail-to-rail gaps between 0 and 1 inch,
- 3) frequent contacts between the rail and the inner wheel flanges of certain wheels,
- 4) apparent misalignment of the drive wheels, and
- 5) railhead wear, loose rail shims, and broken and taut ground cables.
The G&H report concluded that the rail motion is due to a flanging action of the wheels, caused by a
( combination of the off-center crane operation and a
j Ravisien: 1 l
Pcg3 21 ef 42
,m RESULTS REPORT
(,) -
ISAP VI.b e
l (Cont'd) j
)
5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont't) misalignment of the drive wheels. Using data previously available from KRANCO, and a postulated model of the unloaded crane operating off center, G6H "
calculated that radial forces due to flanging were on the order of 10 kips, while the tangential force of the drive wheels is on the order of 4 kips (Reference 9.25).
The G&H report concluded that a practical improvement would be the use of rail splice bars to make a continuous circular rail. The splice bars would permit j thermal and pressurized containment building expansion, i while maintaining the rail gap within controlled limits. The use of the rail splice would not eliminate the wheel flanging and railhead wear, but would mitigate, if not completely relieve, the rail movement problem.
Third Party Investigations (g)
'v' From April 1985 to June 1985, the third party was actively involved in evaluating the rail motion for likely causes and corrective action suggestions. In late June 1985 the third party documented its findings (Reference 9.26).
The third party found that:
- 1) All 14 rail segments show signs of movement,
- 2) The crane does not always rotate about the exact center of the rail circle, and the center of the crane drive wheels do not always align with the center of the rail, (These conditions, acting in combination, tend to develop forces causing wheel flanging that typically occurs on the inboard face.)
- 3) Rail head wear is typically found on the inboard face, and
- 4) Measured rail movements are too large to be caused by ambient levels of thermal expansion.
I
(_
{
Ravisien: 1 Pcg3 22 of 42 RESULTS REPORT
~
ISAP VI.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd)
The conclusion of the third party was that the rail motion is a result of the flanging action of the wheels. This flanging action produces both radial and tangential loads due to the wheel / rail contact. While the rail clips effectively resist the radial forces, the tangential loads are resisted only by the four friction clips and the rail creep restraint; these have proven to be inadequate to prevent the circumferential rail motion.
Polar Crane Rail Movement Measurement In September 1985, Southwest Research Institute (SwRI) performed inspections to obtain detailed information for the Project on the magnitude of the wheel loads and the associated rail motions. On the basis of that information, SwRI prepared the written report of the measurement program (Reference 9.5).
l
( The SWRI report identified that: '
- 1) with friction clips and load cells removed, it was possible to close an original 3/8 inch gap between two rail segments within five complete 360* rotations of the crane, and i
- 2) a load cell assembly used to measure wheel i loads experienced a cumulative tangential wheel load of 42 kips with the friction clips tight and 46 kips with the friction clips loose.
The test served to confirm the rail movement and q provided data that was subsequently used in estimating loads that might be encountered by the redesigned rail clips and splices. I 5.3.2 Rail Attachment Design Modifications Upon assuming design responsibility in October 1986, SWEC initiated a review of earlier G&H proposed design modifications and completed a design to correct the problems of rail movement and to increase the capability of the rail clips to withstand postulated seismic and post-accident conditions. Figure 4 shows
} ,
Revision: 1 Pass 23 of 42 RESULTS REPORT ISAP VI.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) modifications that are intended to maintain rail and alignment and limit rail end gaps to a maximum of one inch while accommodating the maximum crane. thermal growth. This maximum condition occurs when the crane thermal growth exceeds the combination of rail thermal l
growth plus containment pressure and thermal growth resulting from post-pipe rupture environments. This requires that the flexibility of the outboard rail clips be maintained, but allows the inboard clips to be substantially strengthened.
Rail Clips The Polar Crane Seismic Reanalysis concludes that the original rail clips did not meet the design criteria when subjected to SSE loads.
Design Change Authorization DCA-31296, Revision 2, provides a modification to the hold-down clip design by i
/
replacing the old clips on the inside of the rail with heavier clips capable of resisting che rail overturning forces at the base of the rail for all applicable loading combinations. These heavier rail clips will be installed on 24 inch centers and welded )
to the girder. The old inboard clips will not be removed unless required because of construction considerations (Figure 4). The old inboard clips are not relied upon in the current design calculations.
Shims between the rail and the girder will be welded in place to prevent the possibility of shim movement that would affect the contact between the rail clips and the rail flange.
The inboard rail clips will provide the primary seismic l
load restraint, and maintain the rail position during normal operating conditions. The existing outboard j rail clips will not be modified, i SWEC Calculation 16345-EM(S)-002-CZC (Reference 9.27) confirms the adequacy of the new clip design. The l analysis describes and analyzes the mechanism by which the new inboard rail clips will restrain the rail from moving as the rail absorbs seismic loads occurring in the outboard direction. The existing rail clips on the O outside of the rail can sustain the loads to which they are subjected without modifications.
r
t
- Ravisien: 1 Peg 2 24 of 42 RESULTS REPORT
. ,f (j _
ISAP VI.b (Cont'd) '
5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) 1 Rail Splice Bars To resolve the problem of circumferential rail movement due to crane oper:ation and the associated gap formation between rail ends, rail splice bars were designed.
Design Change Authorization DCA-25031, Revision 4 describes the splice bars that Pall be installed to connect adjacent rail ends (Fig.re 4). Two 34-inch-long splice bars with a cross-sectional area of 3.97 square inches each will be used at each splice. The splice bars are bolted to the web of the rail with three through bolts on each side of the splice. These bolts are torqued to provide a friction connection designed to withstand the largest axial rail loads resulting from crane operation, without slippage. The slotted holes in the splice bars and the diameter of the bolts are so dimensioned that a maximum gap of approximately one inch can develop between rail ends
()
7s due to thermal expansion before the bolts engage at the ends of the slotted holes. The splice bars will also maintain a close alignment of the rail ends. In the installation of the splice bars, the rails will be repositioned to obtain a 1/4 inch installation gap.
However, if the rail cannot practically be moved, where ;
the existing gap is less than 1/4 inch, the rail will i be cut. The gap will be filled with a section of rail where the existing gap exceeds 1/4 inch.
The splice bars will provide for a continuous loop so that the individual rail segments will restrain each !
other from moving beyond the one inch gap allowed for '
in the design. The splice bars also provide a mechanism to distribute the seismic rail overturning loads over a rail splice to clips of a neighboring rail segment. This was demonstrated and analyzed in the Polar Crane Seismic Reanalysis. SWEC Calculation 16345-EM(S)-003-CZC (Reference 9.28) was prepared to confirm the splice bar-rail attachment design.
The only mechanisms by which the moving crane can exert axial loads to the rails chat may result in circumferential movement of the rails are 1) by accelerating or decelerating the crane or 2) by radial flanging of wheels against the head of the rail and the
(]
\/
associated potential circumferential creeping movement
Rzvision: 1 P:gs 25 of 42 RESULTS REPORT
)
'\ _- '
ISAP VI.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) of the rail. It is unlikely that circumferential movement of the rails will result from acceleration or braking of the crane (assuming no axial force component due to wheel flanging), because the capacity cf the rail clips and girders and the friction between the rail and girder exceed the maximum acceleration or braking force able to be transmitted from the wheels to the rail. However, should wheel flanging occur due to normal self-centering of the crane, any additional axial force components would be resisted by the splice bars; thus maintaining the required rail position.
The movement resulting from the passage of several wheels, or from several rotations of the crane, can accumulate, and this accumulation can explain the observed formation of gaps. Southwest Research Institute measurements (Reference 9.5) confirmed this
,~ movement and indicated that the maximum axial load in a
( )j rail during crane operation was 46.5 kips. The AISC Specification, Section 1.3.4, recommends a longitudinal force of 10 percent of the maximum wheel loads to be applied at the top of the rail. For the crane this corresponds to an axial load in one rail of 55 kips.
The capacity of the splice with two splice bars, as limited by the capacity of the friction connection of the splice bar, is 89.6 kips (Reference 9.28).
Removal of Existing Rail Attachment Components The existing inboard rail clips will not be removed, except where they may interfera with the installation of the new rail clips. The fcur friction clips associated with each rail segment will be removed; however, the clip holders will be removed only if there is interference with an intended location of a new rail clip. The rail creep restraints will also be removed.
The ground wires and cadwelds will be removed at all rail joints.
5.4 Ceneral Inspection of the Runway 3ystem A general inspection of the crane rail and the runway system was performed in November 1985 by the Project under Traveler CE-85-2876-8902 (Reference 9.6). This inspection was Ih overviewed by the third party. The purpose of the general G'
9
Revision: 1
- Pego 26 of 42 RESULTS REPORT
( .
\,,/ _- ISAP VI.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) inspection was to examine the rail and related components that could have been affected by the rail motion and the upper bracket connections where changes in the gap size had been observed. The girder seat connections were to be addressed later, and thus were not included in the general inspection.
A visual screening of the girder seat connections was performed by SWEC in November 1986 as described in Section 5.2.1.
Four findings were documented. One finding was that rail wear exists to some degree on all 14 rail segments and that some minor spalling of the rail head had occurred. This observed finding was considered as minor in nature and no further action was required.
However, three types of unsatisfactory conditions were observed and documented for other inspection attributes.
These are:
I 1) In 9 locations the upper bracket connections had k-gaps that were different than documented in the previous as-built inspection that formed the basis for the G&H evaluation of the acceptability of gaps in these connections.
- 2) In 5 locations the rail ground cables were completely broken or had partially broken strands.
- 3) In 10 locations the rail leveling shims were loose or damaged.
The ground cables will be removed at the rail ends and the rail leveling shims will be tack welded in place at the same time as the new rail clips and splice bars are installed.
Inspection of upper bracket connections will be performed as part of the Maintenance and Surveillance Program discussed below.
5.5 Review of the Maintenance and Surveillance Program Requirements Related to the Crane Support System The maintenance and surveillance program reviewed by the I third party was presented in the Comanche Peak Steam Electric Station Mechanical Maintenance Manual, Containment Polar Crane Inspection, Instruction No. KHI-317, Revision 1. The review 1
- - - - - - - - _ - _ _ _ _ - _ _ _ _ _ _ _ _ - - _ _ _ _ _ ___ _ \
R vision: 1 Pege 27 of 42 RESULTS REPORT 7
s _-)
~ ISAP VI.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd) concentrated on the then-existing procedures for the maintenance and surveillance of the runway system. It was found that Instruction No. MMI-317 addressed only the crane structure and its mechanical and electrical components. The only reference to the runway was found in Section 5.4.1.c. and this appears to address only the rail.
It was determined by the third party that it would be prudent to conduct an inspection of the runway system and that the existing Instruction No. MMI-317 should be revised to include requirements for such inspection.
Acceptance criteria item were recommended by the third party to TUGCO. Revision 2 of KHI-317 was issued in July 1986 (Ref erence 9.18) and the third party reviewed it and concurred with it (Reference 9.29).
Instruction No. MMI-317, Revision 2 is based on the G&H-designed runway system. The Project has committed to
('_
\m -
- revising the procedure and issuing it as MMP-317 to include the SWEC runway system design (Reference 9.30). The procedure, in combination with requirements that the Project has also committed to be included in an Electrical Maintenance Procedure EMP-343, will assure that an adequate maintenance and surveillance program is in place.
5.6 Out-of-Scope observations In STIR CPRT-S-005 (Reference 9.31), SWEC reported the completion of a seismic analysis of the crane truck assemblies that determined that the crane truck assembly components, except the wheel bearing housing support bolts, were within allowable stress limits for the specified load conditions.
Based on a preliminary analysis, SWEC initiated a design change (DCA 49099, Rev. 0) to replace the bolts with higher strength bolts to assure that the stress levels are within allowable limits. This effort is part of the SWEC review of L j
the KRANCO seismic qualification report for the crane rather i than part of ISAP VI.b.
5.7 Summary of DIRs Concerns related to five different aspects of the polar crane runway system were identified and documented on eleven DIRs f~' (see Table 2):
b) o
R: vision: 1 i
Pega 28 of 42 '
RESULTS REPORT (D
V -
ISAP VI.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd)
- 1. Gaps observed at the girder seat connection and the implications to structural considerations (Observation E-0269), <
l I
- 2. Discrepancies in the design calculations (Observations E-0019, D-0091, D-0092), )
l
- 3. Circumferential movement of the rails resulting in gaps between rail ends (Unclassified deviations E-0270, E-0991, E-1274),
- 4. Lack of design calculations for the rail, rail restraints, the girder and its attachment to the support brackets (Unclassified deviations D-0093, D-1055, D-2261),
- 5. Inconsistency between the crane purchase specification (
and the FSAR (Deviation D-2475). {
One of the initial concerns associated with the polar crane issue involves gaps existing at the girder seat connection.
The implications of these gaps were evaluated, and it was determined that these gaps do not cause a condition adverse to the design. As such, this discrepancy does not constitute a deviation and was, therefore, closed as an observation. ,
The discrepancies identified in design calculations reviewed i by the third party (Item 2 above) were determined to be minor I in nature; i.e., they did not adversely affect the calculation results. Accordingly, these discrepancies do not represent design deviations, such that the associated DIRs were closed as observations.
Normal opsration of the polar crane necessitates a smooth l rail surface on which the wheels travel. Irregularities in j this surface may give rise to dynamic loadings that could {
jeopardize the ability to retain a lifted load adequately. '
The large gaps observed between rail ends (worst case approximately three inches) created a situation its which satisfying the above requirement was questionable. The gaps are a result of a failure to design the rail restraint i adequately to prohibit circumferential movement. {
Additionally, the rail restraints were inadequate for the seismic loading. This failure to design adequate rail O restraints was due to determining the related loads V incorrectly and failing to follow AISC design requirements.
Accordingly, the DIRs associated with Item 3 above are design s deviations. '
Ravision: 1
~Paga 29 of 42 RESULTS REPORT
' \s [ ')/- ~ ISAP VI.b (Cont'd) 5.0 1
IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd)
The three DIRs associated with Item 4, which identify the lack of design calculations for the polar crane support system components, were determined to be design deviations because of the failure to provide design calculations or other design basis documentation for components of a load path of the polar crane, which is a Seismic Category I structure.
During the review process, a discrepancy was found between the seismic load requirements specified in the crane purchase specification (Reference 9.32) and the FSAR (Table 17A-1).
The FSAR states that the crane must be designed to retain the rated load during and after an SSE, while the specification requires that the crane carry only its own deadweight during the SSE. TU Electric has clarified that the specification represents their intended requirements. Accordingly, TU Electric has initiated an FSAR Change Request (Reference 9.33) to correct this discrepancy, which.is a design deviation. The DIR has been closed, based on the FSAR amendment.
fT A safety significance evaluation is required for the DIRs
(_,/ identified in Items 3, 4, and 5, because they are design deviations.
5.8 Safety Significance Evaluation The concerns expressed in four of the eleven DIRs were classified as observations; therefore, they have no safety significance.
The remaining seven DIRs were determined to be deviations.
Due to the uncertainty associated with performance of a safety significant evaluation and the need for corrective action it was decided not to investigate the potential safety significance of these deviations further and to proceed directly to corrective action. Therefore, these seven DIRs are unclassified deviations.
The root cause and generic implications of these seven unclassified deviations are addressed in Section 5.9 below.
The Project also originated three Significant Deficiency Analysis Reports (SDARs) on polar crane related issues:
CP-86-60 addresses the gaps at the girder seats. This
() concern was resolved by the SWEC evaluation of the girder seat connection. The report submitted by letter
- R3 vision: 1 Pega 30 of 42 RESULTS REPORT (3 i i
\~~/ -
ISAP VI.b (Cont'd)
.5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd)
TXX-6544 (Reference 9.34) concluded that, based on the results of the SWEC evaluation of girder seat connection the item was not reportable under the provisions of 10CFR50.55(e).
CP-86-61 reported three broken ground wires, two partially dislodged shims and two f ailed cadwelds at groundwire attachments to rails, indicating rail movement. The report submitted by letter TXX-6092 (Reference 9.35) concluded that the issue was not reportable under the provisions of 10CFR50.55(e).
CP-86-62 addressed a misinterpretation of load cases provided by KRANCO and consequent failure to consider these loads properly in the original design calculations. Based on the SWEC Polar Crane Seismic Reanalysis, the Project concluded that the issue is not reportable under the provisions of 10CFR50.55(e).
() 5.9 Root Cause and Generic Implications As discussed above, there were seven DIRs classified as '
unclassified deviations. In accordance with CPRT Program Plan requirements, the root cause and generic implications of these unclassified deviations were evaluated.
The unclassified deviations fall into two groupings: those associated with a lack of design basis documentation, and those associated with the design of the rail restraints and the rail movement resulting from the inadequacies of the restraints.
I Given that the crane and its support system is Seismic Category I, it is a matter of industry practice that quantitative design basis documentation in the form of calculations or test data be developed to verify their design {
adequacy. No such information was available for several aspects of the design process prior to TRT (see Section l 5.1.1). The operating performance of the rail and associated restraint hardware suggests that certain design details (e.g.,
rail clips) were inadequate and that had proper design evaluation been completed, the design could have been corrected and the problem of rail movement and inadequacy of rail clip design avoided before the polar crane and support
- [' system was installed.
(
l l
l
R; vision: 1 Pag 2 31 of 42
,-ss RESULTS REPORT ISAP VI.b (Cont'd) 5.0 IMPLEMENTATION OF ACTION PLAN AND DISCUSSION OF RESULTS (Cont'd)
The evaluation of the design performed during implementation l i
of this ISAP demonstrated that the original rail clip design concept was inadequate to carry rail overturning loads resulting from a postulated seismic event, as well as to preclude excessive rail movement during normal operations.
Later attempts to remedy the rail movement by the addition of friction clips and rail creep restraints were unsuccessful.
It is apparent that G&H either underestimated or did not 3 quantitatively consider rail design loads for normal operation. In summary, the original rail clip design was inadequate. Dependent on the magnitude of the rail loads the later modifications might have been adequate if the hardware
{
had been properly sized to carry the design loads. It is '
possible that G&H may have developed the series of design modifications by iteration to be proven through subsequent operating experience.
Collectively, this situation represents a failure to r'% quantify important design input parameters adequately (e.g.,
( rail loads) and to factor this information into both the original design process and later remedial efforts. With respect to the history of operational problems with the rail and the continuing evolution of the rail restraint designs over the years, the Project also failed to investigate operational problems adequately and to take appropriate corrective action.
The root cause of these design problems is determined to be weaknesses in the G&H design process and a failure to comply with certain design control requirements of 10CFR50, Appendix B as amplified by ANSI N45.2.11. The design-related problems identified within this ISAP are similar in nature to those identified within the CPRT Design Adequacy Program (DAP). DAP-identified issues, as well as other design-related items, are subject to remedial actions under the comprehensive Corrective Action Program (CAP), being conducted by TU Electric. Given the breadth and depth of the CAP, generic implications associated with this ISAP are considered to be addressed.' Accordingly, no further corrective action under this ISAP is warranted.
The seven DIRs are closed on the basis of the corrective action taken by the Project.
O O
~
Rsvision: 1 Page 32 of 42 RESULTS REP 0FT (3
k-- -
ISAP VI.b (Cont'd)
6.0 CONCLUSION
S The review of the history of the runway system leads to the conclusion that early weaknesses in the design and installation of the runway system were corrected over a period of time, with the exception of the clips and circumferential rail movement. Normal operation of the crane itself has been acceptable and smooth after problems with the rail-wheel misalignments were corrected.
The visual inspection conducted by SWEC identified the worst case girder connection gaps and the location of the associated bearing areas at the those worst case girder seat connections. These data and the bounding analysis confirmed the adequacy of the support and attachment design, regardless of the magnitude and location of visible gaps, and the conclusion that no further action or rework at the girder seat connections was necessary.
The' investigations by ;.he Project, as well as by the third party, confirmed that the earlier attempts to control the circumferential movement of the crane rails were inadequate. It was concluded that f-^- the rail splice design proposed by G&H and finalized by SWEC will
(, correct any problems of misaligned rail ends at the joints. The analytical work by SWEC supports a conclusion that the splice bars will be structurally adequate to limit the gaps between rail ends to the prescribed maximums without unduly restricting the capability of the rail support system to allow for expansions and contractions resulting from temperature variations and postulated accident conditions.
The general inspection did not identify any unsatisfactory conditions that are not corrected by the design modifications or repair activities. ;
}
Finally, it was concluded that the inadequacies in the earlier l maintenance and surveillance program will be corrected by the forthcoming issuance of the procedure MMP-317 and the issuance of EMP-343.
7.0 ONGOING ACTIVITIES An FSAR Change Request was issued that requests a change to the FSAR to clarify the lifted load / seismic load combinations (Reference 9.33). Nuclear Licensing, Dallas, requested, and {
received, additional justification (Reference 9.36) . After review, j Nuclear Licensing intends to submit the request to the NRC in a
[ forthcoming FSAR amendment in late 1987. l l
l l
-___D
Revision: 1
' Pego 33 of 42 RESULTS REPORT o )
()' _
ISAP VI.b (Cont'd) i 4
7.0 ONGOING ACTIVITIES (Cont'd)
Installation of the modified rail attachments and splice bars is underway for Units 1 and 2.
The Containment Polar Crane Mechanical Inspection Procedure, MMP-317, will be revised by October 1,1987 to include the design modifications implemented in the program described above and the Electrical Inspection Procedure, EMP-343, will be issued by October 1, 1987 to provide the electrical inspection requirements. The inspection program will be conducted before polar crane usage at each refueling outage.
8.0 ACTION TO PRECLUDE REOCCURRENCE OF THE ISSUE The modifications to the crane rail attachments, i.e., splice bars and rail clips, in combination with. proper operation of the crane and compliance with the maintenance and surveillance requirements of HMP-317 and EMP-343, is expected to prevent reoccurrence of this issue. The surveillance program will detect any unexpected or
( incipient problems in time to take any necessary corrective action.
9.0 REFERENCES
9.1 Safety Evaluation Report, Supplement 8, NUREG 0797, Related to the Operation of Comanche Peak Steam Electric Station Units 1 and 2, Docket Number 50-445 and 50-446, February 1985 9.2 DALM-209, R. E. Holloway to J. J. Moorhead, " Reactor Building Polar Crane Support System, Shimming of Non-Contact Girder Seat, Upper Bracket Connection and Seismic Connection".
November 18, 1977 9.3 USNRC Letter to TUGCO, Docket: 50-445/82-11, 50-446/82-10 December 13, 1982 9.4 SWEC Calculation, 16345-EM(S)-008-CZC, Revision 1 9.5 Final Report, " Polar Crane Rail Movement Measurements.
Comanche Peak Steam Electric Station", Southwest Research Institute, December 1985 9.6 Polar Crane General Inspection Traveler CE-85-2876-8902, October 31, 1985, with TSG-13727, November 24, 1985
. () 9.7 " Manual of Steel Construction", American Institute of Steel Construction Inc., Seventh Edition, 1970 >
0
Revision: 1
. Page 34 of 42 RESULTS REPORT ISAP VI.b (Cont'd)
9.0 REFERENCES
(Cont'd) 9.8 " Specifications for Electric Overhead Traveling Cranes", Crane Manufacturers Association of America, Inc., C.M.A.A.
Specification //70, 1975 9.9 TERA-213, " History of Unit 1 Polar Crane Installation, Operation, and Inspection Involving Girder Seat Connection and Rail Motion", August 31, 1987 9.10 GHF-1805, R. E. Hersperger to J. J. Moorhead, " Polar Crane Support System", July 26, 1977 9.11 BRF-7404, H. C. Dodd, Jr. to J. T. Merritt, Jr., " Polar Crane Bracket Connections", November 8, 1977 9.12 USNRC Notice of Violation, Docket: 50-445/82-11, July 7, 1982 9.13 Nonconformance Report, M-82-894, July 7, 1982 9.14 USNRC Notice of Violation, Docket: 50-445/84-08, July 26, O
1984 9.15 GTN-68522, R. E. Ballard, Jr., to J. B. George, " Evaluation of Increased Shim Gaps at Seismic Restraints for Polar Crane Supports", February 24, 1984 9.16 GTT-10372. R. E. Ballard to J. B. George, "As-Built Polar Crane Side Bearing Plate Gaps", June 4, 1984 9.17 GTT-10457, R. E. Ballard to J. B. George, "As-Built Polar Crane Side Bearing Plate Gaps", August 3 .984 9.18 CPSES Maintenance Manual MMI-317. Revision 2, July 7, 1986
" Containment Polar Crane Mechanical Inspection" 9.19 TWK-11467, R. E. Heim to H. R. Rock, "(Reference) 2323-S1-0515, S2-0515", October 5, 1979 9.20 GTT-4352. H. R. Rock to J. T. Merritt, Jr., " Unit 1 and Unit 2 Polar Crane Rails", October 24, 1979 9.21 USNRC Letter to TUGCO, Docket: 50-445/50-446, November 29, 1984 9.22 CPSES Preoperational Test Procedure, Polar Crane,1CP-PT-81-03
~ Revision: 1 Page 35 of 42 RESULTS REPORT l \
\~ / -.
ISAP VI.b (Cont'd)
9.0 REFERENCES
(Cont'd) 9.23 CPSES Maintenance Manual MMI-317, Revision 1, December 14, 1983, " Containment Polar Crane Inspection", transmitted by TSG-12,780 9.24 SWEC Calculation, 16345-EM(B)-001-CZC, Revision 4 9.25 Final Report, " Polar Crane P. ail Study", G&H, May 14, 1986 9.26 Letter W. J. Hall /V. J. Mcdonald to C. Mortgat, " Crane Supports and Rails", June 26, 1985 9.27 SWEC Calculation, 16345-EM(S)-002-CZC, Revision 1 9.28 SWEC Calcult.r.Lon, 16345-EM(S)-003-CZC Revision 2 9.29 Speed Lett9r R. W. White to M. Osterman, "TRT VI.b - Polar Crane Mechtnical Inspection", July 24, 1980 9.30 TCF-87441, J. J. Kelley to J. Arros, " File Compilation for
. -w). ISAP VI.b; Polar Crane", August 6,1987 9.31 STIR-CPRT-S-005, Revision 1, " Issues Concerning Structural Adequacy of the CPSES Reactor Building Polar Crane", July 17, 1987 9.32 Specification No. 2323-MS-34, " Containment Polar Cranes" 9.33 NE-6284 "FSAR Change Request" April 1, 1987 9.34 TXX-6544 CPSES, Docket Nos. 50-445 and 50-446, Polar Crane Girder and Girder Support Shim Gaps, SDAR: CP-86-60 (Final Report), June 29, 1987 9.35 TXX-6092 CPSES, Docket Nos. 50-445 and 50-446, Polar Crane Restraints, SD AR: CP-86-61 (Final Report), November 17, 1986 ,
l 9.36 NE-11496, R. T. Jenkins to D. R. Woodlan, " Comanche Peak Steam l Electric Station Polar Crane", August 31, 1987 i
,m-1 1
a
Revision: 1
. Page 36 of 42 RESULTS REPORT ISAP VI.b (Cont'd)
Figure 1 Runway Support Brackets & Girder Plan
~
Rail
, a m . m D' '
1 Crane
. Runway 2 - Grder
- 1.=l .- y
) . .
I A
Crane Runway Support Bracket 14 Rail Segrnents on 28 Crane '
su "Aa T ' Containment Liner 28 an Runway Support Brackets 9 A- SECTION "A-A" 65'-0 to center #ne rail
[
m l
RAll & CRANE RUNWAY GIRDERS PLAN VIEW O
______m_____________________ _ _ _ _ _ _ _ _
Revision:: 1
. Page 37 of 42 RESULTS REPORT I O _ ISAP VI.b (Cont'd)
I Figure 2 Runway Support / Girder Connection -
(
i l
1 Seismic Bradet !
I Upper Bracket Shims Grder Seismic (2 places)
Restraint Attachment 1
Seismic Bracket l
[ d Grder Radal Restraint ment (2 per girder) r Upper a t Support
's' Bracket g ,
.s ,4
, < /N j q Crane Runway Grder (28 total)
Solo Plate 1 irder S6at Shims 1 (2 places) 5 r%
/ 's Crane Runway l Support Bracket (28 total) l 1
i l
Rsvisien: 1 Pega 38 of 42 RESULTS REPORT O ' ISAP VI.b (Cont'd)
Figure 3 Original Rail Clip Assemblies With Added Friction Clips and Rail Creep Restraints Conlerof RailSegrnent
.1 Friction Clip 8 Hekler '
(4 per segment) j
./
/ , # e* Rail Creep Restraint (1 per segmerW)
/
r
\ / f Rail Clip (30 per segment) / _
p/
l/
p d
./sO
~ ~ '
Gap between Rails 3/8 Installed Gap round Wire l
l 3
i
'Revi81on: 1 Page 39 ,f z, RESULTS REPORT i
( ISAP VI.b (Cont'd) l t
Figure 4 Revised Rail Clip Assemblies & Splice Bars New C5p (WCat)
(
1/4" installed Gap l
D ft, Nu- 0 -
Washer (6 each) 0 f 0
O SPEce Bar
/ ,
New Clip s
f (Weal)
/
Existin
( %gCEp rd) nnn$($a,ay C
( y f
O I l
l
Rsvision: 1
. Pege 40 of 42
.,.s .
RESULTS REPORT
'/ -
ISAP VI.b (Cont'd)
Table 1 Polar Crane Support System Design and Installation
. ITEM NUMBER DESCRIPTION Bridge Travel: 360' Bridge Speed : 50 ft/ min Design Life : 40 years Operating Cycle: Two to four weeks each year for 40 years.
Top of rail elevation = 950'-7" Top of steel seat elevation for crane runway support bracket =
947'- 4 1/2" +/- 1/4" Applicable Specifications:
) 2323-SS-10 Reinforcing Steel 2323-SS-11 Cadwelding Rebar 2323-SS-14 Containment Liner Steel 2323-SS-17 Miscellaneous Steel 2323-SS-20 Seismic Criteria for Equ'ipment Design 2323-MS-39 Containment Polar Cranes Applicable Drawings:
KRANCO 7523-CL-1 Clearance Drawing KRANCO 7523-3 Bridge Assembly KRANCO 7523-6 Bridge Wheel Assembly Bostrom- 2386-7 Polar Crane Girders Bergen G&B 2323-SI-0515 Polar Crane Support Details Sheet 5 O
v
- Rsvision: 1 Pcgs 41 of 42 RESULTS REPORT ISAP VI.b (Cont'd)
Table 1 (Cont'd)
ITEM NUMBER DESCRIPTION Containment Environmental Conditions:
Ambient LOCA Temperature 120*F 280*F Pressure O psis 50 psig Containment wall diameter expansion --
2.4 inches l l
Bridge Weight = 593.5 kip ;
Trolly Weight = 416.5 kip Electrical We = 16.0 kip l Crane Weight = 1026.0 kip (513 tons)
V !
l i
I 1
i O
-d
. Rsvision: 1 Pcg2 42 of 42 TABLE 2 p
SUMMARY
OF RELATED DIRs DIR # SUBJECT CLASSIFICATION Inadequate Design Calculations i
D-0019 Calc. SRB-109c, Set 3, Rev. 6 Modeling Error & Assumption Observation f
{
D-0091 Calc. SRB-109c, Set 3. Loads Not j On KRANCO Drawing Observation j D-0092 Calc. SRB-109c, Set 3, Uplift Flange Observation Based on Plastic Sect. Mod.
j 1
D-0093 Calc. SRB-109c, Set 3, No. Cales. '
For Sole Plate Bolts Unclassified Deviation D-1055 Polar Crane - Missing Cales. for Unclassified Girders and Rail Restraints Deviation D-2261 Polar Crane System - No Cales, for Unclassified Rail and Rail Restraints Deviation O Rail Movement & Attachment E-0269 Polar Crane Shims Improper Observation E-0270 Gaps In Polar Crane Rail Unclassified Deviation l E-0991 Rail and Support System Design Process Unclassified Deviation E-1274 Polar Crane Rail Motion Restraint Unclassified Deviation Inconsistent Design Criteria D-2475 Polar Crane - Design Load Requirements Unclassified Deviaeion O
t.
I t________--_-----._ - - - -