05000333/LER-2009-005
| James A. Fitzpatrick Nuclear Power Plant | |
| Event date: | 04-20-2009 |
|---|---|
| Report date: | 06-22-2009 |
| Reporting criterion: | 10 CFR 50.73(a)(2)(i)(B), Prohibited by Technical Specifications |
| 3332009005R00 - NRC Website | |
BACKGROUND
The James A. FitzPatrick Nuclear Power Plant uses eleven (11) two-stage Target Rock Safety / Relief Valves (S/RV) for emergency pressure relief. These valves are located inside the primary containment, and relieve pressure from the main steam lines to'the torus. The valves are manually actuated once during plant start-up to demonstrate their ability to open. They are not subsequently tested during the plant operating cycle since actuation would result in an unnecessary transient and operator challenges.
During each refueling outage approximately one-third of the S/RV main body assemblies and all eleven of the pilot assemblies are removed and replaced with vendor tested and certified components. The main bodies and pilots that are replaced are sent to a vendor facility for testing, refurbishment, and certification. The test results for pilot assemblies removed in September 2008, during Refueling Outage 18, identified 5 S/RVs that would not have opened within the Technical Specification (TS) setpoint tolerance of 1145 psig +/- 3% (1110.7 — 1179.3 psig).
EVENT DESCRIPTION
As-found testing was performed on the eleven Main Safety / Relief Valves (S/RV) pilot assemblies removed in September 2008, during Refueling Outage 18. The testing was conducted between April 9, 2009 and April 17, 2009 by Wyle Laboratories. The TS setpoint for each SRV is 1145 psig +/- 3%. During the initial lift test, five of the eleven pilot assemblies failed to open within the allowed setpoint tolerance of +/-3% (1110.7 to 1179.3 psig).
The following table summarizes the test results for the 5 pilot assemblies that failed the initial lift test.
Over Pilot First Second Acceptance PressureIn-service Serial Test Test Range AnalysisLocation Number (psig) (psig) 1111-1179 Limit 1195 psig 02RV-071B 1237 1192 1163 Unsat Under 02RV-071C 1047 1184 1156 Unsat Under 02RV-071E 1080 1187 1170 Unsat Under 02RV-071F 1238 1227 ---- Unsat Over 02RV-071L 1193 1245 1148 [� Unsat Over As shown in the above table, subsequent tests passed the acceptance criteria of +/-3% for four of the five pilot assemblies. The most probable cause for the high lift setpoint on the four pilot assemblies that were retested was determined to be corrosion bonding between the pilot valve disc and seat. The most probable cause for the failure on the fifth pilot assembly was determined to be significant pilot valve seat leakage. Due to the seat leakage additional steam pressure would have been required to overcome the leakage and lift the S/RV. As shown in this table, pilot 1238 was not retested. This was due to the stated failure mechanism being significant seat leakage. The seat leakage was draining the pressure header of the test equipment such that repeat testing was not feasible.
TS LCO 3.4.3 requires nine operable S/RVs when in Modes 1, 2 or 3. Specifically, the LCO states: "The safety function of 9 S/RVs shall be OPERABLE." Since five of eleven pilot valves exceeded the allowable setpoint tolerance, this report is being made under 10 CFR 50.73(a)(2)(i)(B), "Any operation or condition which was prohibited by the plant's Technical Specifications...
As described above, all eleven S/RVs in the plant had recently rebuilt and certified pilots installed during the refueling outage and were opened during the plant start-up to verify their Operability. Therefore, the currently installed S/RVs are OPERABLE.
EVENT ANALYSIS
The S/RVs provide overpressure protection for the Reactor Coolant Pressure Boundary (RCPB) as required by the ASME Boiler and Pressure Vessel Code. S/RV pilots actuating at pressures higher than the required setpoint may be significant if adequate overpressure protection is not available. The RCPB Overpressure Analysis is performed each fuel cycle, based on the worst case anticipated transient with nine S/RVs opening at an analyzed upper limit pressure of 1195 psig, and two S/RVs out of service.
Three of the five failed S/RV pilots opened within the analyzed upper limit of the overpressure analysis.
Therefore, nine of the eleven S/RVs would have opened within the upper limit of the overpressure analysis and the overpressure analysis remains a bounding analysis.
Although the electric lift system installed in 2000 is not credited in the RCPB Overpressure Analysis, it was OPERABLE throughout the operating cycle. This system uses a pressure switch. to energize solenoid valves which supply pneumatic pressure to assist in overcoming corrosion bonding. Based on industry experience, this electric lift system compensates for corrosion bonding, and was available to mitigate the effects of corrosion bonding on the four S/RV pilot assemblies that exhibited this failure mode.
The high as-found test for S/RV pilot 1238 was not attributed to corrosion bonding. This S/RV Pilot was determined to have failed due to significant seat leakage that would have required additional steam pressure to overcome the seat leakage and lift the valve. Due to the effect of the seat leakage on the test equipment this pilot assembly was tested only once.
The safety significance of this event is considered low and does not decrease the effectiveness of plant barriers providing safety to the public.
CAUSE OF EVENT
The most probable cause for four of the five high out of tolerance pilot setpoints was determined to be corrosion bonding between the S/RV pilot disc and seat [Cause Code B]. With a bond forming between the pilot disc and seat, more pressure is needed to raise the pilot disc off its seat. Since the normal balance of pilot assembly spring force and steam pressure force necessary to lift the pilot disc corresponds to the nominal setpoint of the S/RV, the pilot disc to seat bond results in a higher pilot lift setpoint.
Corrosion bonding is a known phenomenon caused by an oxygen rich environment in the pilot assembly, due to the radiolytic breakdown of water into hydrogen and oxygen. Oxygen accumulates in the area of the pilot disc because he pilot assembly is a high point on the main steam [SB] line.
A contributing cause for corrosion bonding of the pilot disc to seat may be temperature gradients across the pilot assembly, related to the SRV insulation and ventilation airflow. The installation of insulation on the Target Rock S/RVs and redirecting ventilation flow away from the S/RVs has proven to be beneficial in the industry.
The fifth SRV was determined to have failed due to seat leakage. Additional steam pressure would have been required to overcome the leakage and lift the S/RV. The seat leakage identified in pilot 1238 is most likely a result of performing the lift test during plant start-up. There is extensive industry experience with pilot valve leakage after the lift test.
EXTENT OF CONDITION
All of the SRVs are susceptible to setpoint drift due to pilot disc to seat corrosion bonding. This is a recurring industry issue that has been the subject of both Nuclear Regulatory Commission (NRC) and Boiling Water Reactor Owner's Group (BWROG) generic assessments. Based on the known industry wide issues with the two-stage Target Rock S/RVs FitzPatrick has implemented the following industry recommendations:
1. Installed Stellite 21 discs in all of the eleven S/RVs pilot assemblies during refurbishment at the vendor facility; 2. Installed the electric lift system recommended by the BWROG; 3. Installed enhanced insulation on the S/RVs; and 4. Redirected ventilation air flow away from the S/RVs
FAILED COMPONENT IDENTIFICATION
Manufacturer: Target Rock Corporation Model Number: 7567F-10 NPRDS Manufacturer Code: T020 NPRDS CompCnent Code: Valve FitzPatrick Component ID: 02RV-071B, C, E, F, & L
CORRECTIVE ACTIONS
Corrective Actions Completed Prior to this Report:
1. Installed enhanced insulation on pilot assemblies.
2. Redirected ventilation to limit cooling effect.
3. Replaced pilot assemblies with recently refurbished, tested and certified assemblies.
The effect of the lift SetpointS for these S/RVs being out of tolerance was analyzed and the results of this analysis show that Reactor Pressure Vessel (RPV) overpressure protection and nuclear plant safety were not adversely affected. Consequently, the safety significance of this event was minimal.
This event did not result in a safety system functional failure as defined by NUREG 1022 and NEI 99-02, Revision 5.
SIMILAR EVENTS
- 1. JAF LER-07-001 "Safety Relief Valve Setpoint Drift," August 6, 2007.
2. JAF LER-05-002 "Safety Relief Valve Setpoint Drift," June 6, 2005.
, 3. JAF LER-03-002 "Safety Relief Valve Setpoint Drift," October 16, 2003.
4. JAF LER-01-005 "Safety Relief Valve Setpoint Drift," August 17, 2001.
5. JAF LER-99-003 "Safety Relief Valve Setpoint Drift," March 16, 1999.
6. JAF LER-98-002 "Safety Relief Valve Setpoint Drift," April 9, 1998.
REFERENCES
1. JAF Condition Report CR-JAF-2009-01439, Root Cause Analysis Report, Five of the eleven pilots failed as-found testing (testing high out of tolerance).
2. JAF Condition Report CR-JAF-2007-02108, Root Cause Analysis Report, Seven of ten SRV pilots failed as-found testing (testing high out of tolerance).