05000336/FIN-2010006-01
From kanterella
Jump to navigation
Jump to search
Finding | |
---|---|
Title | Broken Jacket Water Banjo Bolt Adversely Impacted EDG 3B Operability |
Description | An unresolved item (URI) was identified because additional information from Dominion and additional NRC review and evaluation is needed to assess the existence of a performance deficiency and its associated characterization (i.e., more than minor, and whether the issue constitutes a violation).During the conduct of 3B EDG routine testing on February 11, 2010, a significant JW system leak occurred at a JW fitting to one of the 14 EDG cylinders. These fittings are referred to as banjo bolts due to their physical configuration resembling a bolt through the body of a banjo. On July 22, 2009, Dominion initiated CR 343051 to address minor JW leakage from the No. 13 cylinder on EDG 38. Dominion estimated the leak rate at approximately 60 drops per minute and determined that it did not have the potential to impact EDG operability. Dominion closed the CR to WO 53102270827. On September22, 2009, operators tagged out EDG 38 for preventive maintenance on the service waterside of the heat exchangers (WO 53102241548). Maintenance completed the planned work on the EDG and operators completed their pre-job briefing for EDG postmaintenance testing. As operators were clearing tags and aligning the EDG for testing, maintenance called to report that during the performance of EDG minor maintenance under WO 53102283391 (to check the leak tightness of No. 13 cylinder banjo bolt), they had discovered that the gasket appeared crushed or the JW fitting could be possibly cracked. They recommended that an immediate repair be pursued. Since the JW banjo bolt tightness check was performed as minor maintenance and not planned into the work window, there were no contingency parts on hand and a corrective maintenance work order was not ready in case of scope expansion. The emergent failure required draining the JW system and resulted in extending the EDG 38unavailability beyond the original planned unavailability (although still within the technical specification allowed outage time). On September 22, 2009, maintenance repaired the JW leak by replacing a degraded banjo bolt on No. 13 cylinder using the original banjo bolt WO 53102270827. Maintenance documented an unanticipated failure of the broken banjo bolt in the WO package. Operations and maintenance supervision reviewed and closed WO 53102270827 with no additional actions taken. On February 11, 2010, operations noted excessive JW leakage from No.3 cylinder during the EDG 3B monthly test, immediately declared the EDG inoperable, performed a controlled shutdown of the EDG, and initiated CR 368610. The team walked down EDG38 shortly after it was shut down and noted that operations had made an appropriate operability decision based on amount of JW that spilled on the floor and the magnitude of the JW leak rate with the EDG shutdown. Dominion determined that the JW leak was from a cracked banjo bolt. The cracked banjo bolt resulted in approximately 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> of unplanned unavailability on EDG 38. Maintenance replaced the banjo bolt on NO.3cylinder and operations declared the EDG operable on February 12 following postmaintenance testing. On February 22, 2010, maintenance replaced the banjo bolts on all 14 cylinders on EDG 38, resulting in approximately ten more hours of EDG unavailability. On February 23, 2010, preliminary results from a magnetic particle inspection of the removed EDG 38 banjo bolts revealed seven additional cracked bolts (CR 369856). On February 23, 2010, maintenance replaced all the banjo bolts on the redundant EDG 3A to address the extent-of-condition. Dominion\\\'s initial review of the banjo bolts removed from EDG 3A did not identify any degraded bolts similar to those removed from EDG 38.The team noted that Dominion took prompt and appropriate corrective actions following the emergent banjo bolt failure on February 11; however, the team identified that Dominion had not initiated a corrective action CR in September 2009 when they had identified the first failed banjo bolt. The team noted that this represented a missed opportunity to evaluate the deficiency within Dominion\\\'s CAP, and may have precluded the emergent EDG unavailability in February 2010. Specifically, Dominion procedure PIAA-200, Corrective Action, Attachment 1, listed examples of conditions that require a CR, several of which were applicable to the unanticipated failure of the banjo bolt ,including 1) deficiencies or adverse conditions identified during performance of work, 2)a component failure that is outside of what would normally be expected, and 3)documentation of equipment failures. The team identified that Dominion did not initiate anew CR for the increased JW leakage that potentially impacted EDG operability or for the failed bolt in September 2009, did not re-open and re-screen the July 2009 CR (CR343051), and did not initiate a CR to perform a Maintenance Rule (MR) functional failure evaluation for the banjo bolt failure. The team noted that the failure to initiate a CR for the failed banjo bolt was a missed opportunity because Dominion proactively addressed other JW leaks on EDG 3B during an additional planned unavailability in December 2009 that required draining the JW system. If Dominion had evaluated the banjo bolt failure within their CAP, they may have inspected a sample of banjo bolts and/or proactively replaced all the banjo bolts on the 3B EDG during the December work window. On February 24, Dominion initiated CR 369962 to perform a MR evaluation for the banjo bolt failure discovered in September 2009. Based on the team\\\'s concerns, Dominion initiated CR 370566 for not identifying the degraded JW banjo bolt condition in the CAP in September 2009 and to evaluate their work order documentation review process to address potential generic concerns in this area. The team determined that the degraded condition identified in September 2009 (the broken banjo bolt) was unanticipated and represented an operability concern in contrast to the relatively minor JW leak identified in July 2009. Also the team was concerned that the failure to document the September 2009 failure and take actions to prevent recurrence could have allowed the February 2010 failure during surveillance testing. However, the team concluded that additional information is needed to fully evaluate and characterize the potential performance deficiency. An unresolved item is an issue of concern about which more information is required to determine if a performance deficiency exists, if the performance deficiency is more than minor, or if the issue of concern constitutes a violation. Therefore, this issue will be treated as an URI. Information necessary to complete the NRC\\\'s review is as follows: The failure mechanism of the banjo bolts, including common cause(s); Dominion\\\'s assessment of EDG 3B prior operability (Le., prior to the February 11monthly test), including the associated reportability determination; Confirmation of maintenance history for banjo bolts on both EDG 3A and 3B(Le., preventive maintenance such as torquing, repairs for leaks, replacement, etc.); Assessment regarding the extensive degradation of banjo bolts on EDG 3B (9 out of14) vs. none on EDG 3A; and Dominion\\\'s assessment/communication regarding 10 CFR Part 21 applicability. Upon availability of the above information, additional NRC review will be required to independently assess Dominion\\\'s associated causal analyses for the issue, and determine the appropriate characterization. Specifically, the NRC will assess 1) whether the issue was reasonably within Dominion\\\'s ability to foresee and correct prior to February 2010,2) the banjo bolt failure mechanism, 3) EDG fault exposure, and 4) any associated violations. (URI 05000423/2010006-01, Broken Jacket Water Banjo Bolt Adversely Impacted EDG 3B Operability) |
Site: | Millstone |
---|---|
Report | IR 05000336/2010006 Section 4OA2 |
Date counted | Mar 31, 2010 (2010Q1) |
Type: | URI: |
cornerstone | Mitigating Systems |
Identified by: | NRC identified |
Inspection Procedure: | IP 71152 |
Inspectors (proximate) | T Moslak J Caruso D Jackson S Pindale B Haagensen B Welling P Presby J Krafty S Shaffer J Johnsonp Kaufmand Jackson S Pindale J Schoppy J Krafty |
INPO aspect | |
' | |
Finding - Millstone - IR 05000336/2010006 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Finding List (Millstone) @ 2010Q1
Self-Identified List (Millstone)
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||