05000458/LER-2009-001
| Docket Numbersequential Revmonth Day Year Year Month Day Yearnumber No. 05000 | |
| Event date: | |
|---|---|
| Report date: | |
| 4582009001R00 - NRC Website | |
On January 14, 2009, an investigation of operational practices related to the standby liquid control (SLC) (**BR**) system concluded that for the period of March 14, 2003, to October 28, 2008, a seismic event could have rendered the system incapable of performing one of its design safety functions as credited in the station's accident analysis. This condition resulted from an inadequately evaluated change made to a surveillance test procedure that allowed water to remain in the system's test tank (**TK**). The test tank is not seismically analyzed when full of water. This event is reportable in accordance with 10CFR50.73 as operations prohibited by Technical Specifications and a loss of the safety function of the SLC system.
BACKGROUND
The SLC system is a safety related design feature installed to provide a means of shutting the reactor down in the event of a failure of the reactor control rod drive system. The system contains redundant pumps powered from the emergency diesel generators that would be used to inject a sodium pentaborate solution into the reactor when directed by operating procedures. The solution is contained in a tank that can supply either or both of the pumps.
A separate 250-gallon tank is built into the system to serve as a suction source of clean water for performing periodic surveillance testing of the pumps and the associated motor-operated valves. The test tank has steel supports that elevate it approximately 16 inches off the floor. The test tank is designed to remain intact in the event of the design basis seismic event, assuming that the tank is empty.
When RBS implemented License Amendment No. 132 (Alternative Source Term) in 2003, the SLC system gained a new design function related to operations following a postulated loss of coolant accident (COCA). In order to mitigate the release of radionuclides into the primary containment atmosphere in the post-LOCA environment, the SLC system is assumed to be used for pH control of the suppression pool. The analysis assumes that the SLC system is initiated at a certain time in the event, and that the borated solution will leave the reactor vessel through the postulated break in the coolant system piping. The solution would drain to the suppression pool where it would provide a buffering effect in maintaining the pH above a value of 7.
Prior to the implementation of the Alternative Source Term amendment, suppression pool pH control was not a design function of the system.
CAUSAL ANALYSIS and IMMEDIATE CORRECTIVE ACTIONS The investigation of this event found that, in 1992, surveillance test procedures were revised to remove the requirement to drain the test tank upon completion of surveillance tests prior to returning the system to service. The review of the proposed procedure change did not surface the fact that the test tank is not designed to withstand seismic loading when it is full of water.
To determine whether the filled condition of the test tank had any adverse effects on the operability of the SLC system, an analysis was performed to identify the likely failure effects. It was found that the tank could possibly fall in two different directions due to failure of the supports. In the worst case scenario, cables attached to a junction box mounted in the immediate vicinity of the test tank outlet valve could be damaged due to impact. The limit switches on the test tank outlet valve are necessary to satisfy an "open" interlock with Division 1 and 2 outlet valves on the main storage tank (i.e:, the storage tank outlet valves will not open unless the test tank outlet valve is fully closed). Breakage of the cable from the "CLOSED" limit switch would block any opening signal to the storage tank outlet valves. Since it could be postulated that both outlet valves would be affected, this scenario would render the SLC system incapable of performing its function.
When the status of the SLC test tank was originally questioned in October 2008, the tank was drained as a conservative measure since it was not clear what potentially adverse effects were posed by keeping the tank full. It was the investigation of that question that determined the specific details contained in this report.
CORRECTIVE ACTIONS TO PREVENT RECURRENCE
An administrative tracking mechanism has been put into place for the affected surveillance test regarding the draining of the test tank. Guidance for draining the SLC test tank was incorporated into the system operating procedure.
Further procedure revisions are being developed. This action is being tracked in the station's corrective action program.
PREVIOUS OCCURRENCE EVALUATION
A review of events reported by River Bend Station since January 2004 found no similar conditions.
SAFETY SIGNIFICANCE
The past condition of storing water in the SLC Test Tank had no actual impact on nuclear safety as there were no seismic events or other events requiring SLC system response. However, during the time period March 14, 2003, through October 28, 2008, the postulated test tank failure could have rendered the SLC system unable to respond following a postulated LOCA in conjunction with a seismic event. The resultant lack of suppression pool pH control could have resulted in aerosol particulate iodine (cesium iodide) deposited in the suppression pool tore-evolve and become airborne as elemental iodine. Prior to the implementation of the Alternative Source Term amendment, suppression pool pH control was not a design function of the system.
RBS has not performed a dose consequences evaluation with re-evolution of iodine. However, an iodine re-evolution study performed at a plant of similar design has concluded that the impact on doses due to re-evolution is negligible. Thus, while RBS has not performed a plant specific dose consequences evaluation with `respect to iodine re-evolution, it is reasonable to conclude that the impact on calculated dose from iodine re-evolution would be minor. Correspondingly, it is reasonable to conclude that given the margin between the RBS Alternate Source Term calculated doses and 10CFR50.67 limits, the minor change in dose from iodine re-evolution would not result in exceeding 10CFR50.67 limits.
(NOTE: Energy Industry Component Identification codes and system identification codes are annotated as (**XX**).)