05000482/LER-2009-005

LER-2009-005, Loss of both Diesel Generators with all fuel in the Spent Fuel Pool

Docket Number
Event date:	10-22-2009
Report date:	2-6-2013
Reporting criterion:	10 CFR 50.73(a)(2)(v)(B), Loss of Safety Function - Remove Residual Heat
4822009005R01 - NRC Website

Control Room annunciators 20D, DG NE01 Trouble, and 20B, DG NE01 UV or UF, for the "A" Diesel Generator (DG) [El IS Code: EK] actuated. At 1208 CDT the "A" DG was taken out of service for troubleshooting of abnormal alarm indications.

At 1739 CDT on 10/22/2009, a Notification of Unusual Event (NUE) was declared due to the loss of the "A" DG with the "B" DG out of service for refueling outage maintenance. The NUE was delayed due to the licensed operator failure to recognize that "All" Modes in the Emergency Action Level Classification scheme is effective at all times. The delay in the NUE has been entered into the WCGS corrective action program. At the time of the event, the reactor vessel was defueled with all fuel assemblies located in the Spent Fuel Pool. There are no Technical Specifications (TS) requirements for the DGs when defueled. The TS defines Mode as "A MODE shall correspond to any one inclusive combination of core reactivity condition, power level, average reactor coolant temperature, and reactor vessel head closure bolt tensioning specified in Table 1.1-1 with fuel in the reactor vessel.

Power to the safety related busses were being supplied by offsite power.

The direct cause of the event was the speed switches in the starting circuit of the A DG had actuated due to the failure of the capacitor on the DC input feed to the annunciator power supply. This allowed higher levels of AC noise to be applied to the DC feed to the speed switches. The high AC noise was more than the capacitors on the DC feed to the speed switches could filter out. The AC noise actuated the switches.

The source of electrical noise on the DC supply circuit was traced to the annunicator power supply within the emergency generator gauge board panel. The annunciator power supply and speed switch were replaced on the A DG and the new switch calibrated.

The "A" DG was returned to service at 0738 CDT on 10/23/2009. The NUE was exited at 0740 CDT on 10/23/2009.

Further investigation determined that the root cause of the speed switch actuation was a wiring error of the speed switch. The control cabinet supplier did not wire the speed switch per the recommendation of the speed switch manufacturer. As a result the speed switch signal wire was not protected from ambient electromagnetic interference.

This event is reportable in accordance with 10 CFR 50.73(a)(2)(v)(B), (C) and (D) as an event or condition that could have prevented the fulfillment of the safety function of structures or systems that are needed to: (B) Remove residual heat; (C) Control the release of radioactive material; or (D) Mitigate the consequences of an accident. Specifically, the guidance in NUREG-1022, Rev. 2, states: "Both offsite electrical power (transmission lines) and onsite emergency power (usually diesel generators) are considered to be separate functions by GDC-17. If either offsite power or onsite emergency power is unavailable to the plant, it is reportable regardless of whether the other system is available. GDC-17 defines the safety function of each system as providing sufficient capacity and capability, etc., assuming that the other system is not available.

ROOT CAUSE:

The root cause of the false speed switch actuation was a wiring error of the speed switch. The control cabinet supplier did not wire the speed switch per the recommendation of the speed switch manufacturer. As a result the speed switch signal wire was not protected from ambient electromagnetic interference.

Laboratory testing identified that the termination point of the speed switch signal wire shield had a significant impact on the noise being received (induced) on the signal wire by electromagnetic frequencies (EMF). When wired per the speed switch manufacturer document, the shield provided a significant reduction in the amplitude of received (induced) noise on the signal wire. The reduction of EMF induced noise on the signal wire was so significant, it has been concluded that had the speed switch been wired per the manufacturer document, no false actuation event would have occurred.

CORRECTIVE ACTIONS:

The speed switch termination point has been corrected.

A PM activity was established to have the AC ripple voltage measured at the DC input to the annunciator power supply every three months. If the measured AC ripple voltage reaches a value of 500 milivolts, a work request will be generated to have the power supply replaced.

SAFETY SIGNIFICANCE:

The Outage Control Center performed a shutdown risk assessment on 10/22/2009 after the loss of the DG. The shutdown safety risk condition for the plant was Risk condition 2. Risk condition 2 indicates a moderate risk. There is a reduction in the equipment available for satisfying a shutdown safety function because of electrical power resources and decay heat removal.

The entire core was in the Spent Fuel Pool (SFP) and the SFP water inventory was full. The plant- operating mode was de-fueled. Switchyard activities were not in progress. Power was available from the offsite power sources. The SFP time to boil was 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br />.