05000336/LER-2002-002
| Event date: | 04-19-2002 |
|---|---|
| Report date: | 06-18-2002 |
| Reporting criterion: | 10 CFR 50.73(a)(2)(iv), System Actuation 10 CFR 50.73(a)(2)(IV)(A) |
| 3362002002R00 - NRC Website | |
1. Event Description On April 19, 2002, with the plant operating at approximately 99.5 percent power, a turbine [TRB] trip resulted in an automatic reactor [RCT] trip when a temperature switch [TS] in the Stator Water Cooling System tripped on a high value. The temperature switch in the Stator Water Cooling System was found with a trip setpoint value approximately 14 degrees Celsius (°C) lower than the acceptance criteria found in the calibration procedure.
This incorrect trip setpoint was just above the normal system temperature, so that on April 19, 2002, an increase in the stator water cooling temperature tripped temperature switch 63T-72. The increase in temperature of the stator cooling water was primarily due to the increase in generator power that was occurring at the time of the trip.
Additionally, to a lesser extent, the increase in stator cooling water temperature was affected by an increase in the circulating water inlet temperature. This increase in inlet temperature was caused by an environmental condition that allows hot water from the plant discharge point to sweep back to the intake structure. The increase in water temperature at the intake structure [NN] affected the Service Water system directly, and the Turbine Building Closed Cooling Water system [KB] indirectly.
The purpose of the Stator Water Cooling System is to remove heat in the main generator [GEN] stator produced during the electrical generation process, and supply cooling water to the static rectifiers ERECT] used in the excitation system. It also functions to purify the cooling system, maintaining a very low level of electrical conductivity.
In March , 2002, during refueling outage 14, the stator water outlet temperature switch was calibrated and found to have a setpoint of 80.4°C, within the accepted range of 80-82°C. The reset point was recorded as 50.2°C. There is no acceptance criteria given for the reset point, so it was left as found. Following the plant trip on April 19, 2002, the temperature sensor was checked and found to have a setpoint of 66°C with a reset value of 60.1°C. The sensor was re-calibrated to an acceptable setpoint of 82°C with a reset value of 79.4°C and the plant was re- started the following day.
The cause of the incorrect setpoints has not been determined. Following the plant trip, repeated checks of the calibration, both as-found and as-left were completed. The calibration checks were completed using both a dry- method and a water bath. Both methods gave similar results. No indication of equipment failure could be identified and the plant was restarted the following day using the same sensor.
Possible failure modes were investigated. Vibration has been a concern on the stator water cooling skid with respect to pump/motor reliability, and steps have been taken in the past to help reduce these vibrations. Most recently, during the last refueling outage, epoxy was injected under the pump [P] baseplate to dampen resonant frequencies. In the past, vibration has not had the type of impact on the setpoints that were experienced in this event, and discussions with the vendor indicate that setpoint drift of this magnitude is possible, however highly unlikely. The vendor would have to do a thorough inspection of the switch to determine whether vibration caused this magnitude of setpoint drift.
Another possibility for equipment failure is that the six foot long capillary tube that connects the temperature sensing bulb to the bourdon tube is not routed in conduit [CND]. Because it is exposed, there is a possibility that the capillary tube could come into contact with a nearby hot pipe and therefore have an adverse effect on the sensor readings. However, the capillary tube has historically not been affected in such a manner to cause a trip signal. Therefore, no plausible equipment failure scenarios could be identified that would produce a turbine trip signal.
Although no credible failure mechanism for the temperature switch could be determined, the very lack of any conclusive evidence forces a consideration to be made that some undiscovered failure of the temperature switch may have been the cause of the low setpoint.
Additionally, the event may have been a result of a human error in calibrating the temperature sensor. The Stator Winding Cooling Water System Calibration procedure covers the calibration of 23 instruments in this system. Only one of the instruments covered by this procedure generates a trip signal for the turbine, the stator cooling outlet temperature. There is no caution statement and only a small notification in the procedure to alert the user to the trip potential of the instrument. The other trip signal in the system, the low system pressure sensor, is calibrated using a separate procedure, Stator Cooling System Setup.
The Stator Winding Cooling Water System Calibration procedure provides minimal guidance for performing the calibration. It provides no guidance on key points such as the value of the reset setpoint, the size chuck to use in the King Nutronics dry method calibration unit, or the rate at which the temperature is to be increased or decreased during the calibration. While this information is considered skill of the craft within the Instrumentation and Control department, the technicians vary widely in their interpretations of the values that should be used.
Testing using the King Nutronics calibration unit along with a temperature switch similar to the one installed in the stator cooling system was performed to test calibration techniques. The testing was performed using different combinations of common errors experienced during calibration of a temperature sensor including incorrect aluminum chuck sizes on the calibration unit, and different heatup and cooldown rates. The testing proved that incorrect setpoint and reset values matching those found on the switch following the reactor trip, could be obtained.
This event is being reported pursuant to 10 CFR 50.73(a)(2)(iv) as an event that resulted in an automatic reactor scram.
2. Cause The cause of this event is indeterminate. The presumptive causes are either an equipment failure or insufficient controls that led to calibration errors.
3. Assessment of Safety Consequences The purpose of the Stator Water Cooling System is to remove heat in the main generator stator produced during the electrical generation process, and supply cooling water to the static rectifiers used in the excitation system. It also functions to purify the cooling system, maintaining a very low level of electrical conductivity.
Following the reactor trip, main steam reheat valves, 2-MS-2A and 2-MS-2B, did not close as expected and had to be manually closed, resulting in an unexpected RCS cooldown. Auxiliary Feedwater [BA] automatically initiated to compensate for a low level in the Steam Generators [SG] caused by shrinkage. Required safety equipment functioned as expected and this event is considered to be of low safety significance.
4. Corrective Action Following the reactor trip, the temperature switch was tested and re-calibrated to the proper setpoints prior to plant restart. To prevent recurrence the stator coolant temperature switch from both Millstone Unit Nos. 2 and 3 will be replaced. The Unit No. 2 device will then be sent to the vendor for a failure analysis. In addition, separate procedures will be developed for secondary plant instruments that are trip sensitive at Unit Nos. 2 and 3.
Additional corrective actions are being addressed in accordance with the Millstone Corrective Action Program.
5. Previous Occurrences None Energy Industry Identification System (EIIS) codes are identified in the text as [XX].