05000316/LER-2010-001
| Docket Number | |
| Event date: | 10-06-2010 |
|---|---|
| Report date: | 12-1-2010 |
| Reporting criterion: | 10 CFR 50.73(a)(2)(iv)(A), System Actuation |
| Initial Reporting | |
| ENS 46311 | 10 CFR 50.72(b)(3)(iv)(A), System Actuation |
| 3162010001R00 - NRC Website | |
Conditions Prior to Event Mode 3
Description of Event
On October 6, 2010, at 0008 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, a valid automatic actuation of the Unit 2 Turbine Driven Auxiliary Feedwater Pump [P] (TDAFP) occurred as a result of two of four Steam Generator [SG] (S/G) levels reaching their low-low level setpoint.
At 0001, in preparation for the start of the scheduled refueling outage, the Unit 2 reactor [ROT] was manually tripped in accordance with the Power Reduction normal operating procedure. The East and West Motor Driven Auxiliary Feedwater Pumps had been started immediately prior to the reactor trip in order to maintain flow to the S/Gs per procedure.
Following the reactor trip, the operators completed their immediate actions and adjusted Auxiliary Feedwater (AFW) [BA] flow in accordance with procedure. Transition was then made to the Reactor Trip Response procedure to stabilize the plant.
The first step of the Reactor Trip Response procedure directs the operators to determine if a Reactor Coolant System [AB] (RCS) cooldown is in progress. Operators initially determined that no cooldown was evident based on control board indication. Subsequently, as operators were adjusting AFW flow, a cooldown of the RCS was noted, but the source of the cooldown was not immediately identified. Operators believed that the cooldown was due to high AFW flow, so they began to throttle AFW flow.
At 0008, the TDAFP started automatically as a result of two of four S/G levels reaching the low-low level setpoint. The cause of the cooldown was later determined to be primarily due to steam flow through the Steam Dump system [JI].
Malfunction of the Steam Dump system was determined to have resulted from an out-of-calibration electro-pneumatic transducer (EPT).
The following conditions contributed to the lowering generator water levels. First was operators throttling AFW Flow in an attempt to arrest the RCS cooldown. Second was the loss of mass due to the steam flow through the Steam Dump system.
An additional contributor was level in S/G 4 lowering when the AFW valve for S/G 1 failed to automatically throttle upon receipt of the flow retention signal, diverting flow from S/G 4.
Approximately one minute after the start of the TDAFP, Permissive [69] P-12 cleared. The clearing of P-12 blocks closed the Steam Dump valves. With the Steam Dump valves blocked closed, the primary contributor to the RCS cooldowh was stopped. The TDAFP was stopped approximately nine minutes after it auto-started.
At 0052, per procedure direction, the Steam Dump system was transferred from T-Avg Mode to Steam Pressure Mode.
This removed the out-of-calibration component from the circuit, and the Steam Dump system operated as designed to maintain RCS temperature.
There were no structures, systems or components known to be inoperable at the start of the event that contributed to the event.
Cause of Event
The automatic start of the TDAFP was a result of two S/G levels lowering to the low-low level setpoint.
Levels were lowering due to an RCS cooldown and loss of secondary system mass caused by steam flow through the Steam Dump System. An out-of-calibration EPT in the Steam Dump control system resulted in two Steam Dump valves remaining partially open with no demand signal. This flowpath resulted in sufficient steam flow to the Main Condenser [SG] to cool down the RCS. This cooldown and loss of secondary system mass resulted in S/G levels lowering to the setpoint for automatic start of the TDAFP.
The diagnosis of RCS cooldown was delayed, allowing two S/Gs to reach their low-low level, due to a malfunctioning AFW supply valve for S/G 1. Rather than throttling automatically upon receipt of a flow retention signal, the valve remained full open until operators manually operated
- the valve to throttle flow. This resulted in operators focusing initially on high AFW flow as the source of the cooldown, thus delaying the diagnosis of the steam flow through the Steam Dump system.
Troubleshooting and functional testing of the S/G 1 AFW supply valve and control circuitry resulted in the components and circuitry operating properly. The cause of S/G 1 AFW supply valve operating incorrectly could not be conclusively determined.
Analysis of Event
An assessment of this event determined that it is bounded by the existing accident analysis associated with unplanned reactor trips (i.e., transient) with the main condenser (i.e., ultimate heat sink) available. The change in risk with respect to core damage and large early release frequency due to an out-of-calibration electro-pneumatic transmitter in the Condenser Steam Dump system, and subsequent minor RCS cooldown, have been qualitatively assessed and judged to be no different than any other planned reactor trip with the main condenser available. This assessment is based on the following considerations:
1. The automatic plant responses to Steam Generator water level and the P-12 setpoint, caused by the EPT out-of- calibration condition, functioned as expected. Operators took procedurally directed actions and responded to the transient in an appropriate and timely manner, resulting in a safe and stable plant configuration. Automatic post-trip features functioned dependably with the exception of the S/G 1 AFW valve not throttling closed following the trip. The latter condition is judged to not have had a significant impact on the plant transient.
2. The unexpected minor RCS cooldown, caused by the EPT out-of-calibration condition, does not contribute to the increased likelihood of any initiating events.
3. Neither the EPT out-of-calibration condition nor the S/G 1 AFW valve not moving to its throttled position degraded any system used to mitigate core damage, assure containment integrity, or maintain defense-in-depth and safety margins.
Based upon an examination of the event, the risk significance associated with the RCS cooldown following the planned October 6, 2010, Unit 2 trip is assessed as non-risk significant.
Corrective Actions
Completed Corrective Actions
The electro-pneumatic transducer found out of calibration has been replaced.
Although no cause for the incorrectly operating AFW valve could be conclusively determined, an action was taken to replace the time delay relay in the associated circuitry as this was determined to be the most likely cause. Functional testing following relay replacement identified that the circuitry is functioning properly.
Planned Corrective Actions
An action has been generated to change the Preventive Maintenance activity for the EPT found to be degraded in this event. The Preventive Maintenance activity will direct replacement of the EPT, rather than calibration, if the as-found condition of the EPT is outside a specified tolerance. This is being performed because the EPT has been found to drift out of tolerance sooner if calibrated from outside the specified tolerance.
Ah action has been generated to add the lessons learned from this event to the Pre-Job Brief data base for planned plant shutdown evolutions.
Previous Similar Events
Licensee Event Reports for the past 10 years were reviewed. Below is the only LER identified that was initiated as a result of a similar type of an unanticipated actuation of the auxiliary feedwater system.
In preparation for a Unit 2 refueling outage, operators performed a manual reactor trip of Unit 2 from 22 percent power.
Shortly thereafter, an automatic start of the TDAFP occurred as a result of valid low-low levels in two of four Steam Generators: The reactor trip setpoint had been selected to avoid challenging ESF equipment (i.e., auto start of the TDAFP). As such, the automatic start of the TDAFP was not specifically called out as an expected occurrence after manual reactor trip from 22 percent power.
While this previous event was similar in that the TDAFP started automatically, the previous TDAFP automatic start was attributed to tripping the reactor from a power level of 22 percent rather than due to RCS cooldown caused by equipment malfunction. To address this issue, the procedure was changed to trip from 14 percent.
1.