05000458/LER-2010-004
| Docket Numbersequential Revmonth Day Year Year Month Day Yearnumber No. 05000 | |
| Event date: | 11-07-2010 |
|---|---|
| Report date: | 03-02-2011 |
| Reporting criterion: | 10 CFR 50.73(a)(2)(v)(D), Loss of Safety Function - Mitigate the Consequences of an Accident |
| 4582010004R01 - NRC Website | |
On July 14, 2010, a minor, unquantifiable oil leak was found on the lower reservoir drain plug of the HPCS pump. The leak appeared only as oil sheen around the drain plug. On August 7, the leak was quantified as approximately 1 drop every 3 minutes. A maintenance technician tightened the plug approximately one-quarter turn that day, which had no apparent effect. The leakage rate was determined to be stable, and it was concluded that the pump remained capable of performing its safety function.
The approximate timeline of subsequent activities concerning this event, developed from documentation and from interviews with the operators and maintenance technicians, is as follows. Oil was added to the reservoir on September 13. The HPCS system was operated on September 20 for scheduled surveillance testing, and no increase in the leakage rate was seen.
Oil was again added to the reservoir on October 26. No further oil additions were made until November 7. Twice-weekly inspections of the pump by the operators confirmed that the leakage had not increased. (Operators also perform a general inspection of the pump room each shift, checking for oil accumulation on equipment and other conditions. Those inspections continued to be satisfactory.) On October 29, the operators wrapped an absorbent pad around the plug to eliminate the need to clean up oil around the pump pedestal, and to eliminate the potential slipping hazard. These pads were subsequently replaced three times prior to November 7 (the last replacement was on November 5), and on each occasion, the oil leak rate had not increased.
On November 7, the operator removed the absorbent pad, and found that the leak had increased to a small stream approximately one-tenth of an inch in diameter. The HPCS pump was removed from service, and the oil drain plug was replaced with a new part.
Visual inspection of the removed drain plug assembly identified a circumferential fracture on the threaded portion of the drain plug. The drain plug assembly and 0-ring were then sent to an independent laboratory for analysis.
Laboratory analysis found that the plug fracture surface exhibited a purely ductile failure of a low tensile strength material with significant porosity. The lab postulated that the initial leakage of the drain plug was the result of 0-ring failure. The 0-ring exhibited surface cracks and chips and had developed a permanent set in the radial direction due to its more than 24 years of service. It appeared that an attempt to mitigate the leakage was made by further tightening of the drain plug. The fracture along the head of the plug caused the markedly higher leak rate discovered on November 7.
The pump motor vendor (General Electric) was contacted to provide detailed information about the drain plug material and design of the drain plug assembly and 0-ring. The laboratory analysis determined that the failed drain plug material was zinc alloy die casting, which is consistent with one of the specified materials by GE.
GE also provided service information recommending that low nitrile Buna 0-rings be replaced with medium Buna-N 0-rings containing at least 27 percent nitrile to improve seal reliability and increase the life of the seals. It was also recommended that the 0-rings should be replaced at every other oil change. The laboratory analysis found that the drain plug 0-ring was the recommended 27 percent nitrile material.
That same service information was reviewed by RBS in 1991, and it was agreed to replace the drain plug 0-rings and sight glass 0-rings. The review stated that "A [preventative maintenance] task or multiple tasks will be initiated to track work which will be performed in accordance with the GE SIL." This investigation found that no such tasks were ever initiated.
CORRECTIVE ACTIONS TO PREVENT RECURRENCE
The failed drain plug assembly and 0-ring are common to the upper and lower motor bearing reservoirs of the five emergency core cooling pump motors (i.e., low pressure core spray, three residual heat removal subsystems, and high pressure core spray). The remaining drain plug assemblies are also originally installed components. A schedule has been set for replacing all informed not to apply torque to the currently installed plugs in an attempt to stop any future leakage.
Preventative maintenance tasks will be developed to replace the drain plug 0-rings on a schedule consistent with vendor recommendations.
PREVIOUS OCCURRENCE EVALUATION
There have been no similar events reported by RBS since January 1, 2005.
SAFETY SIGNIFICANCE
Two of three divisions of ECCS are required for the RBS loss of coolant accident analyses.
While HPCS was out of service, Division 1 and Division 2 ECCS systems and the automatic depressurization system were available, and would have met the ECCS performance criteria of 10CFR50.46. The HPCS system was returned to service within the time limit of the Required Action in the plant's Technical Specifications. This event was of minimal safety significance with respect to the health and safety of the public.
(NOTE: Energy Industry Component Identification codes are annotated as (**XX**).)