05000455/LER-2011-001
| Byron Station, Unit 2 | |
| Event date: | 11-17-2010 |
|---|---|
| Report date: | 06-22-2011 |
| 4552011001R01 - NRC Website | |
A. Plant Operating Conditions Before the Event
Unit 2 was in Mode 1 — Power Operations at 100% nominal power.
Reactor Coolant System (RC) [AB]: Normal operating temperature and pressure.
No structures, systems, or components were inoperable at the start of this event that contributed to the initiation or mitigation of this event.
B. Description of Event
On November 17, 2010, during a monthly surveillance run, the Unit 2 "A" (2A) Emergency Onsite Power Supply Diesel Generator (DG) [EK] was operating at full load when the Equipment Operator (EO), assignee to monitor the 2A DG, identified a significant lubricating oil leak from the upper oil cooler. The Main Control Room was immediately notified and the Unit operator began unloading the DG. The EO, at the 2A DG, then tripped the DG by depressing the emergency stop pushbutton. The 2A DG was declared inoperable and Technical Specifications 3.8.1, "AC Sources — Operating" Condition B was entered for one required DG inoperable. Troubleshooting and repair efforts were initiated.
The oil leak source was determined to be coming from the upper lubricating oil cooler at the bolted flange connection between the cooler's shell and the stationary channel head. The as-found bolted torque values were found to be significantly less than the expected value of 110 foot pounds. The loose bolts, and the resulting reduction in clamping force, allowed the gasket to extrude, which caused the oil leak. It was also determined that contributing to the cause of the bolt loosening on the upper oil cooler flange was misalignment between the outlet nozzle flange and lower oil cooler. The aligning and tightening of this lower flange after the upper oil cooler flange was torqued caused the upper oil cooler flange, bolting, and gasket to re-align to a more relaxed neutral position; resulting in the previously torqued bolts to become loose.
The upper oil cooler flange and the outlet nozzle flange had been disassembled during a previous 2A DG work window on January 17, 2010, to repair a leaking tube-to-tube sheet joint. After repairs were completed, a new gasket material was installed and the flange re-assembled. The work instructions provided maintenance personnel with direction to torque each separate mechanical joint to the proper torque value. However, these instructions were not consistent with an Electric Power Research Institute (EPRI) recommendation, for critical and problem joints, to re-torque after 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> to accommodate any unusual high short term relaxation. In addition, the gasket material manufacturer made a similar recommendation.
No leaks were identified during the post maintenance test run of the 2A DG, nor were there leaks observe during subsequent monthly surveillance runs prior to the November 17, 2010, failure. The lower torque combined with the normal 2A DG thermal cycles and vibration, as well as some expected gasket set over time, eventually resulted in the gasket extruding and subsequently leaking during the November 17, 2010.
run. Total run time on the 2A DG from work completion after the January 17, 2010, work until the leak occurred on November 17, 2010, was approximately 64 hours7.407407e-4 days <br />0.0178 hours <br />1.058201e-4 weeks <br />2.4352e-5 months <br />.
An evaluation was initiated to assess the 2A DG TS operability status from January 17, 2010, to November 17, 2010. Based on the results of this evaluation, on January 25, 2011, Exelon Generation Company, LLC (EGC) concluded that on May 19, 2010, the 2A DG would not have been able to fulfill its 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> mission time and therefore was unknowingly inoperable since this date until November 17, 2010. TS 3.8.1, Condition B requires an inoperable DG to be restored to operable status within 14 days, otherwise Condition G requires the Unit to be in Mode 3 within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and Mode 5 within 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br />. Consequently, the condition of Unit 2 not in compliance with Condition G is a condition prohibited by TS and reportable to the NRC in accordance with 10 CFR 50.73 (a)(2)(i)(B).
On June 2, 2010, the 2B DG was declared inoperable due to a failed primary fuse in the voltage potential transformer. The 2B DG was restored to operable status in approximately 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. Since this unplanned event occurred during the timeframe the 2A DG was unknowingly inoperable, this event of both DGs being inoperable is considered an event or condition that could have prevented the fulfillment of a safety function and is also reportable in accordance with 10 CFR 50.73 (a)(2)(v).
The 2A DG was repaired and returned to operable status on November 22, 2010. An extent of condition check of lubricating oil coolers bolted connections found no other loose bolting.
C. Cause/s of the Event The cause was determined to be an inadequate work instructions provided for the January 17, 2010, 2A DG work activities in that it did not contain sufficient directions to re-torque the bolted connections after initial torquing as recommended by EPRI and the gasket manufacturer. This was caused by a lack of a formal process to ensure EPRI documents are systematically reviewed for good practices and then incorporated into maintenance procedures.
D. Safety Significance
There were no actual safety consequences as a result of this condition. Safety bus 241 is normally supplied by off site power via its System Auxiliary Transformer (SAT) 242-1 and was available throughout the time frame that the 2A DG was unknowingly inoperable. Safety bus 242 was availab e and capable of being powered by its offsite power source and would have supplied power to B train components to safely shutdown the plant as designed.
The DGs are designed to provide emergency power to the safety buses in the event of a Loss-0 - Offsite-Power (LOOP). Had a LOOP occurred, the 2A DG would potentially not have run for the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> mission time. The Unit 2 safety bus 241 would have become de-energized. Operating procedure would be utilized to energize bus 241 from Unit 1 safety bus 141. This source of power was available during this time frame.
In the event of a Station Blackout, either one of the two emergency DG for each unit serves as an alternate a-c power source for the opposite unit. The alternate a-c power source is available within 10 minutes of the onset of the Station Blackout event and has sufficient capacity and capability to operate equipment necessary to bring and maintain the station in a safe shutdown condition.
Equipment required for Station Blackout is capable of being powered from a single remaining DG.
The capability for providing power to the blacked-out unit is possible with manual operation of cross-tie switchgear breakers from the Main Control Room.
A probability risk assessment performed for this condition concluded it to be of low to moderate significance.
The June 2, 2010, condition when both the 2A and 2B DGs were inoperable and unavailable is considered of low safety significance. In the event of a LOOP occurring at the time the 2B DG failed, the 2A DG would have started and run for approximately 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> while the 2B was being restored to availability.
E. Corrective Actions
The maintenance procedure governing assembly of bolted connections has been revised to address the assembly of equipment with multiple joints.
The model work orders for DG oil coolers assembly will be revised to require final torque checks after assembly and to include a requirement to ensure torque verification is performed within three months.
The maintenance procedure for assembly of bolted connections will be revised to incorporate EPRI recommendations.
The operating experience review process has been revised to include EPRI documents to ensure good practices are incorporated into EGC's procedures.
Training will be conducted for appropriate maintenance personnel concerning the assembly of multiple joints configurations.
F. Previous Occurrences
Several DG oil cooler flange leaks have been documented in the Corrective Action Program over the past 10 years. The leaks were minor in nature and did not result in the DG being declared inoperable. The repair typically involved re-torquing the flange bolts. No investigations were conducted into determining the cause involving the need to re-torque bolts.