05000254/LER-2007-003
| 'Quad Cities Nuclear Power Station | |
| Event date: | |
|---|---|
| Report date: | |
| Reporting criterion: | 10 CFR 50.73(a)(2)(v)(D), Loss of Safety Function - Mitigate the Consequences of an Accident |
| Initial Reporting | |
| ENS 43793 | 10 CFR 50.72(b)(3)(v)(D), Loss of Safety Function - Mitigate the Consequences of an Accident |
| 2542007003R00 - NRC Website | |
PLANT AND SYSTEM IDENTIFICATION
General Electric - Boiling Water Reactor, 2957 Megawatts Thermal Rated Core Power Energy Industry Identification System (EIIS) codes are identified in the text as [XX].
EVENT IDENTIFICATION
Safety Function Not Met Due to Control Room Emergency Ventilation System Air Filtration Unit Heater Flow Switch Failure
A. CONDITION PRIOR TO EVENT
Unit: 1 Event Date: November 20, 2007 Event Time: 0505 hours0.00584 days <br />0.14 hours <br />8.349868e-4 weeks <br />1.921525e-4 months <br /> Reactor Mode: 1 Mode Name: Power Operation Power Level: 97%
B. DESCRIPTION OF EVENT
On November 20, 2007, at 0505 hours0.00584 days <br />0.14 hours <br />8.349868e-4 weeks <br />1.921525e-4 months <br /> during full power operation, Operations surveillances determined that the Control Room Emergency Ventilation System (CREVS) [VI] (also described as Train B Control Room HVAC), Air Filtration Unit (AFU) [AHU] heaters [EHTR] were not operating as required. The surveillance requirement for Technical Specification (Tech Spec) 3.7.4 is a 10 hour1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> run to be performed once every 31 days with the heaters operating. CREVS was declared inoperable and Tech Spec Limiting Condition for Operation 3.7.4, Action A, was entered which requires the CREVS to be restored to operable status in seven days. This event was reported to NRC on November 20, 2007, per 10CFR50.72(b)(3)(v)(D), (ENS Number 43793), as an event of condition that could have prevented the fulfillment of a safety function, because the CREVS is a single train system required to mitigate the consequences of an accident.
Troubleshooting of the AFU heater indicated that flow switch [FS] FS 0-5795-350 had an open contact [29] that was preventing the heater from energizing and operating as required. All other relays [RLY], switches [IS], and contacts in the AFU heater circuit were found to be operating satisfactorily. By 1625 hours0.0188 days <br />0.451 hours <br />0.00269 weeks <br />6.183125e-4 months <br />, FS 0-5795-350 had been replaced, proper operation of the flow switch and the AFU heaters was demonstrated, CREVS was declared operable, and Tech Spec 3.7.4, Action A, was exited.
Flow switch FS 0-5795-350 is designed to close and allow the heater to operate if the flow through the AFU is sufficient to support the operation of the heater. The heater coils [HCL] require a minimum flow of approximately 1600 cfm (normal flow is 2000 cfm) such that the heater coils do not trip or burn out. FS 0-5795-350 is a Columbus Electric Manufacturing Model RH15-05-D1 flow switch. This flow switch had been replaced and successfully tested on October 24, 2007. The AFU had not been operated from October 24, 2007 until November 20, 2007, when the AFU heater was found not operating.
CREVS is a single safety-related ventilation system that feeds the control room envelope. CREVS contains a single air filtration unit that is designed to pull 2000 cfm of outside air and filters it through a pre-filter, HEPA filter [FLT], heater, and, charcoal adsorber [ADS]. The 2000 cfm is required to pressurize the control room envelope to 1/8 inch water gauge. The AFU heater is specifically designed to reduce the relative humidity of the air traveling to the NRC`FORM 366A(9-2007) PRINTED ON RECYCLED PAPER 4 charcoal adsorbers to less than 70%. A relative humidity (RH) of less than 70% increases the efficiency of the charcoal adsorber in removing radioactive iodide from the air. Maintaining a relative humidity of less than 70% ensures that Tech Spec 5.5.7.c (Ventilation Filter Testing Program) is met, which requires that methyl iodide penetration is less than 0.5%, when tested at 30°C and 70% RH.
Flow switch FS 0-5795-350 is correctly classified and preventive maintenance tasks were being performed correctly per the maintenance template for air handling units. System monitoring and trending data for the AFU heater revealed no signs of degradation. No operating experience information could be found on switches failing within one month of operation. The operating procedures for the CREVS AFU are sufficient and within the design basis for the components. There have been no previous issues with the quality of parts obtained for this application.
Exelon Powerlabs initial failure analysis identified an issue with the open switch contact, in that when testing it would not close when it reached its setpoint on several different occasions. This is an infant mortality issue, given the fact that this was only the second time the switch was required to operate after installation.
C. CAUSE OF EVENT
Flow switch FS 0-5759-350 was installed as part of a preventive maintenance task on October 23, 2007. The flow switch was then successfully tested on October 24, 2007. The AFU had not been operated from October 24, 2007 until November 20, 2007, when the AFU heater was found not operating. During troubleshooting on November 20, 2007, Instrument Maintenance (IM) removed the flow switch sensing line, and upon reinstallation the flow switch contact could be heard going closed. IM was unable to recreate the open contact condition.
Exelon Powerlabs initial failure analysis indicated a problem with the open switch contact which would not close when it reached its setpoint on several different occasions when tested. The resultant failure is attributed to infant mortality, since this switch was only installed for one month and its failure occurred during its second installed actuation. This is the first issue the station has had with this make and model flow switch since its initial installation in 1985.
D. SAFETY ANALYSIS
The safety significance of this event was minimal. Despite the loss of the AFU heater, the charcoal adsorber methyl iodide removal efficiency was sufficient to meet the Tech Spec requirements and there would be no impact to any control room dose calculations or assumptions.
The flow switch used to control the operation of the CREVS AFU heater is a Columbus Electric Manufacturing Model RH15-05-D1. This model switch is not used in any other application at Quad Cities Station. Consequently, impact of this condition with the plant operating was minimal.
E. CORRECTIVE ACTIONS
- FS 0-5795-350 was replaced with a new switch.
- The Exelon Powerlabs failure analysis final report will be reviewed to identify and initiate any additional necessary corrective actions.
�
F. PREVIOUS OCCURRENCES
There were no previous occurrences at Quad Cities Station that involved failure of this flow switch. This is the first issue the station has had with this make and model flow switch since its initial installation in 1985. There was a prior degraded heater event on October 12, 2005 at Quad Cities in which the AFU heater was found degraded. This event is, however, not applicable as a previous occurrence since the AFU heater was found degraded due to issues with Measurement and Test Equipment (M&TE) instruments that were being used to measure voltage and amps on the heater. When two different M&TE instruments were used, the heater was found to be operating properly.
G. COMPONENT FAILURE DATA
The flow switch used to control the operation of the CREVS AFU heater is a Columbus Electric Manufacturing Model RH15-05-D1. This make and model switch is not used in any other application at Quad Cities Station. The Standby Gas Treatment System contains an electric heater that is started based on a flow switch. However, this flow switch is a Rosemount model flow switch.
An EPIX/NPRDS search identified Palo Verde 1 — Failure Number 811 (6/7/2005) Fuel Building AFU heaters failed to operate due to a failed flow switch. This failure is partially applicable to the Quad Cities event in that the switch was the same make and model flow switch as used in the Quad Cities CREVS AFU heater circuit. However, the Palo Verde failure mode was thermal degradation due to age of the switch diaphragm. The Palo Verde switch had been installed for 15 years when the failure occurred. The Quad Cities flow switch had been installed for one month when the failure occurred.
�