IR 05000269/1991037
| ML16148A607 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 01/24/1992 |
| From: | Blake J, Economos N NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML16148A606 | List: |
| References | |
| 50-269-91-37, 50-270-91-37, 50-287-91-37, NUDOCS 9202240114 | |
| Download: ML16148A607 (7) | |
Text
p,9P REG&1 UNITED STATES o
NUCLEAR REGULATORY COMMISSION
REGION II
101 MARIETTA STREET, ATLANTA, GEORGIA 30323 Report Nos.:
50-269/91-37, 50-270/91-37, and 50-287/91-37 Licensee: Duke Power Company 422 South Church Street Charlotte, NC 28242 Docket Nos.: 50-269, 50-270, License Nos.: DPR-38, DPR-47,. and and 50-287 DPR-55 Facility Name:
Oconee 1, 2, and 3 Inspection
!ted De mber 22, 23, 27, 1991 Inspector:
le;(/
N. 'E. _onomof ate Yig ned Approved by:
J./J. &ake, Chief Date Signed teal s and Processes Section ng eering Branch Division of' Reactor Safety SUMMARY Scope:
On December 15, 1991, the licensee informed Region II that a leak had been detected in Unit 3 low, pressure injection (LPI) system suction line which is used for normal decay heat remova The licensee indicated that the crack was located on the 12 inch diameter LPI suction line, down-stream of 3LPI-2 valve, next to a socket weld for the 3/4 inch branch connection for relief valve
- 3LP-25. The crack was located in a section of the LPI system that is isolated during power operations and was not a part of the RCS boundary or part of the ECCS LPI System. The crack was estimated to be approximately 3 inches long, and partly located in the toe of the wel In response to this information, the.inspector conducted an inspection to observe the pipe configuration the location of the crack, to discuss the plans for pipe replacement and/or repair, to observe work in process and evaluate the licensee's performance during thi undertaki Results:
By observation of work in process, discussions with cognizant personnel and document review, the inspector identified the following areas where apparent programatic weaknesses were eviden PDR ADOCK 05000269 G
1. Preliminary failure investigation results indicate the crack was caused by high cycle, low amplitude fatigue associated with ;LPI pump vibrations which had gone undetected until the subject crack develope. Through an apparent lack of on site technical expertise, welding/mechanical engineering during the planning stages of this effort, the initial replacement spool was improperly fabricated and had to be scrapped for failure to meet minimum wall requirement Another spool piece had to be fabricated using a different approach that ultimately proved to.be correct for the applicatio. An apparent lack of. adequate preparation, for providing purge during welding, caused fabrication difficulties in one out of the two 12 inch welds and resulted in numerous repairs requiring three additional days of work before a code acceptable weld could be produce. Because of programatic.changes and recent personnel reassignments, locating and retrieving quality records for replacement materials proved to. be a difficult tas For example, the licensee was able to retrieve quality records for only one of the four replacement components prior to the inspector's departure on December 23, 1991. Two records were faxed to Region II and, the certificate for the 12 inch diameter pipe did not become available until December 27, or the final day of this inspection when the pipe was being welded to the syste The inspector identified these weaknesses/concerns to the licensee prior to and durina the exit intervie On December 31,- 1991 the licensee issued Problem Investigation Report, S/N 3-091-0126 which provided a description of the problem, initial assessment of the root cause, corrective actions to prevent recurrence, possible generic implications for Units 1 and 2 and, proposed resolution Within the areas inspected, violations or deviations were not identifie REPORT DETAILS 1. Persons Contacted Licensee Employees H. B. Baron, Station Manager G. Blubaugh, Technical Support QA B. W. Carney, Mechanical Technical Support Engineer E. Few, Mechanical, Technical Support W. W. Foster, Maintenance Superintendent C. Henson, Technical Support QA J. Kendrick, Shift Manager
- 0. Kohler, NRC License Coordinator
- B. Milsapps, Maintenance Services Manager H. Moore, NDE Technician Other licensee employees contacted during this inspection..included technical, support craft, and operations personne NRC Resident Inspectors P. Harmon, Senior Resident Inspector
- B. B. Desai,.Resident Inspector
a. Background On December 15, 1991 Region II was notified that a leak had been discovered in the Oconee Unit 3 low pressure injection system suction line used for normal decay heat removal. The licensee indicated that the leak was on the 12 inch pipe downstream from valve 3LP-The crack was approximately three inches long and was located in the weld area of the one inch branch connection for relief valve 3LP-25. This branch connection consists of a one inch, half coupling welded to the 12.finch diameter schedule 10S type 304 stainless steel pipe, a short 3/4" diameter stub welded to a flange to connect the subject relief valv The licensee indicated that the 12" diameter pipe section containing the crack could be isolated by valves 3LPI-2 and 3LPI-3, that the section was isolated during plant operation and that it was not a part of the RCS boundary or a part of the ECCS, LPI system. The location of the pipe in question was said to be in the, "Decay Heat Drop Line", in the basement of the Unit 3 reactor buildin Or December 22, 1991, before the pipe was cut and removed from site for investigation, the inspector visited Oconee to inspect the pipe credk under field condition to obSErVE the as-built pipe configuration, to discuss plans for repair/replacement. and to observe oncoine ictivities in this are Visual Inspection By visual inspection, the inspector confirmed earlier reports concerning pipe configuraticn, crack location and mgrpholog Specifically the inspector noted that the crack was at the top of the weld and extended ir both directions, running parallel to the discharne line 'of thE relief valve, on the downstream side of the connecticn. Rust accumulation was observed on the area of the crac Access to the immediate area was rather difficult because cf other piping, instruments arc structural steel interference Cortairation and hidh radiation in the immediate area required respiratory protection and a wet sui Followino this inspection, the pipe section containing the crack was cut, the area of interest was removed from the pipe and subsequently transferred to the hot machine shop where the inspector observed and phctccraphed the crack surfaces on both sides of the pip Followina this work effort, the inspector discussed the licenscee planis for fabrication and installation of the replacement spool p..cC Durino the discussion, the inspector ascertained thaL the subject valve had been replaced on two previews occasions, the lAthst charce bcire early in 1991 when the present valve was installe At that time, the onc inch branch line was replaced with.the present schedule 40, 3/4 inch diameter pip Plans for replacing the 12 inch pipe sectic. and the associated branch connection were as follows:
(1) Replace the existirn 12 inch diameter schedule 10S, pipe section with one made of schedule 40S pipe made of the same materia (2)
Inst&l:i a 3/4" half coupling and a branch connection made of 3/4" diameter schedule 80 pipe (3)
Lower,the valve by 2 5/8."
The licensee indicated that the thicker pipe material would reduce blLraticon loading and increase the fatigue strength of the subject branch connectio Vibration tests were to be performed during start-up to confirm the frequency of the modified branch connection and rionitor line vibrations at that tim Welding and installation of new Pipe This modification was implemented through Exempt Change N OE-4385 and Engineering Calculation OSC-4579 dated December 19, 199 A 10 CFR 50.59 evaluation was performed to consider possible unreviewed safety questions and none were identifid. Following this document review the inspector observed in-process machinino on the replacement spool piec.The inspector noted that the half coupling had beer weldea on the 12 inch diameter replacement spool piece and observed the machinist as he wa.s attempting to center the spool on the lathe in preparation for machining the weld pre The inspector discussed the operation with the machinist and documented material identification numbers for review of quality record Later in the day, the inspector learned that the machined spool piece had been rejected because certain sections in the weld prep area had failed to meet minimum wall requirement The apparent reason for this problem va ettributed to out-of-roundess/distortion in the replacement spool piec.Licensee personnel following this project, discussed the problEm, reviewed the possibilities for repair by welding and decided that it was better to scrap the piece and fabricate a new one. This time the" cceided aoeinst welding out the half coupling on the 12" diameter spocl since that contributed to the distortion proble Instead it was decidcd tc weld only the root of the coupling and leave the remaining weld to be deposited in the field following completion of the large diameter pipe weld Vith these precautions and increased supervision, the replacement spool was fabricated successfull The inspector expressed concern over the apparent lack of adequate review and planning prior to fahricatirc the first replacement piec The inspector stated that welding and mechanical support engineers should have beer mcre activ on this jo[ This would have ensured that each step of the fabrication was well thought out, properly carried out, and most likely would have preverted the need -to fabricate a second replacement spoo On December 23, 1991 the replacement spool was fitted and wElders began to fabricate the new welds, which appear on isometric drawing No. 44 of the LPI syste Work Request N L was issued to document and control work activitie The new welds were fabricated using tungsten inert gas welding process (TIC).
The welding procedure ard welders used on the job were qualified in accordance with AMSE Code Section IX and.the applicable Section XI requirement The controlling construction code was ANSI B31.7, 1969 Edition. The new welds underwent. volumetric/radiographic examination as required by the subject construction code and surface examination to satisfy presservice (PSI)
examination reCuirements of ASME Code Section Y, Within these areas the inspector noted that one of the two 12 inch da. welds (No. 37) was fabricated successfully, without the need for
repairs, and was accepted by radiography on the 24th of Decembe Loss of purge/inert gas atmosphere, during fabrication of the second weld, caused sections. of this weld (N ) to be rejected twic Other fabrication defects resulted. in additional rejections, all of which required a total of *four repairs before fabricating an acceptable weld, late on December 2 On this date, the inspector reviewed the approved radiograph of weld No. 37, and the acceptable sections of weld No. 38. The inspector asked the licensee to reshoot one section of weld No. 37, identified as station 3-4 to verify the presence or absence of an apparent fabrication defect observed on the radiograph during revie The inspector was subsequently notified that a. reshot of the area in question showed no evidence of the apparent indicatio As stated earlier, the inspector, expressed concern over the inability to maintain adequate purge over the time required to weld these two joint The inspector stated that he viewed this problem as a weakness in the area of technical support and as another example of inadequate pre-planning; in this case by welding.supervision who should have planned for adequate purge dams to guard against loss of purge and the aforemention rejection d. Records Review On December 23,- the inspector requested the licensee to provide quality records for.the replacement components for revie The components of interest were as follows:
Iterm Teyp Size Heat N QA Tag Pipe SA-312 3/4" schedule 80 97446 CC003137 Flange SA-182'
3/4" Flange C3950 63693 Coupling SA-1.82 3/4" 3000#
EAP 51198 Pipe SA-312 12" schedule 40 8654046 63772 (.3759)
Filler Metal ER-308 3/32" x 36" PA825 60150 Personnel on site at the time of this request tried unsuccessfully to retrieve the requested record The explanation given for this problem included that key personnel were on annual leave because of the holidays, that revisions to the program eliminated the requirement to have copies of material certifications available to QA inspectors in the field and, that reassignment of cognizant personnel made it difficult to know who to contact for assistance in this are Consequently after it became evident that the subject records could not.be retrieved by personnel on site, the licensee representative telephoned the cognizant individual for assistanc Several hours after the initial request and following the arrival of cognizant individual the licensee produced quality records for only one of the
items requeste At this time the.nspector decided to end the inspection and reoestecd that the subject records be faxed to RegJon II offic Two additional records were received by fax but the certificate for the 12" diameter pipe dio not become available until the 27th of December when the inspector revisited the site to review the radiographs of the new weld The licensee's inability to retrieve the subject records in a timely matter was considered a programatic weakness in this are The inspector stated that these records shculd have been retrieved and reviewed for acequacy prior to releasina the material for installatio Prelimninary Report of Failure Analysis On December 27, Babcock and Wilcox (B&W) Lynchburg, VA provided the licensee with a preliminary report of their failure analysis of the cracked pip In additicn to the report B&W telephoned to.discuss theiE rs-ults w-*th the.licensee and the inspector who was invited to atten B&W-reported that examination of the fracture surface disclosed the crack was C initiated and propagated to the I of the pip That point of origin was placed the center of the crack, along the toe of the wel Their examihrtin of the main part of the fracture, revealed fine fatiue striations with spacings typically less than one micro B&W indicated that the presence of striations on the fracture surface sugaests that the failure mcde was fatiOu Additionally, D&W' sittec that the presence of fire striations indicates the presence of a high cycle, low amplitude. load mechanis B&W concluded tha the crack origin, and lack of corrosicr.ffects; indicates the failure was caused by mechanical fatigu Followirc the B&W report, the licensee issued a Problem Investigation Report (PIR) Serial N,
dated December 31, 199 This report addressed possible generic implications for all three units and proposed resolutions for Units 1 and These rsoGutions provide for vibratior rcecircs and visual inspections to be performed at desionated time Further evaluations of root cause' and generic applicability will be corducted to adequately resolve the aforementioned PI Within the area: inspected violations or deviations were not identifie.
Exit Interview The inspection scope and results were summarized on December 27, 199 with those persons indicated in paragraph The inspector described the areas inspected and discussed irn detail the inspection result Dissenting comments were not received from the licensee.