05000333/LER-2010-001
| James A. Fitzpatrick Nuclear Power Plant | |
| Event date: | 03-22-2010 |
|---|---|
| Report date: | 05-20-2010 |
| Reporting criterion: | 10 CFR 50.73(a)(2)(i)(B), Prohibited by Technical Specifications 10 CFR 50.73(a)(2)(vii), Common Cause Inoperability |
| 3332010001R00 - NRC Website | |
EVENT DESCRIPTION
Safety valves in the residual heat removal (RHR) [ElIS System Identifier: BO] and core spray (CS) [ElIS System Identifier: BM] systems at the James A. FitzPatrick Nuclear Power Plant (JAF) failed to meet In-Service Testing (1ST) as-found acceptance criteria during testing in the 4th 10 year interval conducted in February and March 2010. The most probable cause of the relief valve setpoint failure was the presence of internal binding and/or disc to seat corrosion bonding. The specific components, setpoints, lift pressures, and test dates are in the table below:
SETPOINT 1ST LIFT 211D LIFTCOMPONENT ID COMPONENT DESCRIPTION Test Date(PSIG) (PSIG) (PSIG) 10SV-35A RHR LOOP A SAFETY VALVE 291 - 309 328 328 February 5, 2010 10SV-35B RHR LOOP B SAFETY VALVE 291 - 309 340 335 March 23, 2010 14SV-20A CS PUMP A DISCH SAFETY VALVE 485 - 515 530 530 March 30, 2010 Component testing that falls outside the required action range for IST requires that the component be declared noperable and the applicable Limiting Condition for Operation (LCO) declared not met. For tests required by American Society of Mechanical Engineers (ASME)Section XI that do not meet acceptance criteria, 10 CFR 50.73(a)(2)(i)(B) allows licensees to consider the failure to have occurred at time of discovery unless there is firm evidence to indicate that the condition existed previously. This same logic is applied to testing required by the ASME OM Code. Based on a review of test history for these valves, there is indication that this condition existed prior to discovery.
The event is reportable pursuant to 10 CFR 50.73(a)(2)(i)(B), "Any operation or condition which was prohibited by the plant's Technical Specifications..." because during the time period described, the RHR and CS subsystems were inoperable longer than allowed by the TS completion times without a reduction in MODE. This event is also reportable per 10 CFR 50.73(a)(2)(vii), "Any event where a single cause or condition caused at least one independent train or channel to become inoperable in multiple systems or two independent trains to become inoperable in a single system" designed to remove residual heat or mitigate the consequences of an accident.
BACKGROUND
10SV-35A and 10SV-35B, RHR "A" and "B" Loop Safety Valves 10SV-35A and 10SV-35B are the RHR "A" and "B" loop safety valves [ElIS Component Identifier: PSV]. They are 1" Crosby valves, Model JR-B. They are categorized as ASME Class 2 relief type actuators with IST requirements and are the only two IST program valves in Group 1. The safety functions of 10SV-35A and B are to open in the event of overpressure, such that the RHR discharge piping is protected if pressure exceeds the relief setpoint due to leak-back of reactor water through the low pressure coolant injection (LPCI) [ElIS System Identifier: BO] or due to thermal expansion of the water in the discharge piping. Although they do not have a specific accident mitigating function, they do serve RHR safety-related functions by protecting the RHR system loop integrity.
Originally the preventative maintenance (PM) frequency for these safety valves was every eight years with a 25% grace period because 1ST program requirements dictate that the valves be tested on ten year intervals. However, due to a history of setpoint test failures, the PM interval had been previously reduced to four years. It was believed that by increasing the test frequency the failure rate would be reduced. The ASME requirement is to test 20% of a valve group population every four years A review of test history for the previous 29 years indicates that the increased test frequency did not reduce the magnitude of setpoint drift for as-found testing. The table below summarizes the failure rates obtained since 1981. Replacements due to leakage or failure other than as-found testing were not considered as setpoint failures. All as-found results that were found high were considered failures whereas all as-found results that were low were considered acceptable per OM-1 standards.
Safety valve 14SV-20A is the Core Spray pump "A" loop discharge safety valve. It is a 1-1/2 inch Crosby valve, Model 9811852E. Along with 14SV-20B, it is included in the IST program as a Group 3 test valve. The design function of this valve is to open, in an overpressure condition, such that the CS pump discharge piping is protected.
In 1990 this valve was replaced with a valve containing a stainless steel body. This was done because of a failed as-found test caused by gross internal corrosion buildup in the previously installed carbon-steel body. Since 1990, the new stainless steel valve has been tested a total of four times with a 50% success rate. As with 10SV-35A/B, the ASME requirement is to test a minimum of 20% of a valve group population every four years. Currently, both valves in this group are on a 10 year test interval. The other valve in the group, 14SV 20B, was originally scheduled to be tested in 2016. Due to the failure of 14SV-20A, this test has been rescheduled to June 2010. Based on a review of test history for 14SV-20B and the engineering evaluation performed (see Assessment of Safety Consequences) as a result of the setpoint failure for 14SV-20A, it is reasonable to assume that 14SV-20B is capable of performing its safety function by protecting the CS discharge piping from overpressure. Thus the June 2010 test date is acceptable.
COMPONENT FAILURE RATE NO. OF FAILURES (1990-2010)
EVENT ANALYSIS
ASME 0Mb Code-2003 Addenda to ASME OM Code-2001, Mandatory Appendix I specifies requirements for In-Service Testing of pressure relief devices. It allows the licensee to establish acceptance criteria for the tests required by Appendix I based on the greater of either system or valve design specifications, or +/- 3% of the valve nameplate set-pressure. For the valves being described in this LER, the IST program at JAF has established the valve nameplate set-pressure, +/- 3%, as the acceptance criteria. For the two groups of valves in question, the ASME requirement is to test a minimum of 20% of a valve group population every four years. In accordance with the code requirements, if a valve fails, then the sample population is increased. In the case of Groups 1 and 3, the increased sample size results in testing 100% of the population.
When IST performance data falls outside the required action range, NRC Inspection Manual Part 9900: Technical Guidance on Operability Determinations requires that the component be declared inoperable, the applicable TS LCO declared not met, and the appropriate action statement(s) entered. Since the relief valves were removed from the system at the time of testing, a review of PM history was performed to identify if there were any concerns with past operability that would indicate that the condition existed prior to discovery.
Since 1981, 10SV-35A had experienced a 30% (3 out of 10) failure rate whereas 10SV-35B had experienced a 25% (2 out of 8) failure rate. Based on the evidence presented by the previous failures, it is assumed that the condition existed prior to discovery. The failure of 10SV-35A and 10SV-35B to meet IST acceptance criteria rendered both trains of RHR inoperable. Therefore, LCO 3.5.1 Condition H, "Two or more low pressure ECCS injection/spray subsystems inoperable for reasons other than Condition A" should have been entered.
LCO 3.5.1, Condition H has a Required Action to enter LCO 3.0.3 immediately. Because the condition was not identified during the surveillance interval, a condition prohibited by technical specifications existed for a period of time exceeding the completion time.
10SV-35A and 10SV-35B both have a common failure mode and are both required to be operable in order to maintain operability of their respective trains. Therefore, this event is also reportable pursuant to 10 CFR 50.73(a)(2)(vii), "Any event where a single cause or condition caused two independent trains or channels to become inoperable" in a single system designed to remove residual heat or mitigate the consequences of an accident.
Test history for 14SV-20A was also reviewed. In an as-found test in May 1990 the valve failed to lift due to gross internal corrosion buildup. The valve was then replaced with a stainless steel valve body. Since then, this valve has experienced a 50% (2 out of 4) failure rate. The failure in March 2010 is attributed to internal binding. As such, it is also reportable as a condition prohibited by Technical Specifications and as an event where a single cause or condition caused one independent train in multiple systems to become inoperable.
Engineering personnel performed an inspection of the valve internals for 10SV-35A. The inspection found a slight discoloration on the inlet side of the valve disc and an accumulation of rust material in the discharge piping. The rust was found along the bottom of the pipe and ran all the way to the valve disc and seat surface. The rust had hardened to the disc and seat surface which was likely causing bonding of the two surfaces. This indicated that corrosion bonding (disc to seat) may have been present. 10SV-35B was not available for physical inspection, but based on the as-found testing results, there was no noted change between the first and second lifts. This indicates internal binding was occurring.
An inspection of the valve internals for 14SV-20A was also performed. The inspection did not find any foreign material on the inlet of the valve and no discoloration was noted on the inlet side of the valve disc. Neither did the inspection find any accumulation of rust on the discharge side of the valve. The stainless steel valve exhibited very clean internal surfaces and thus the inspection could not determine what may have caused the setpoint failure. The most probable cause is internal binding as described in EPRI Report TR-105872.
EPRI Report TR-105872, Safety and Relief Valve Testing and Maintenance Guide, was reviewed for input on possible causes for the relief valve lifting high during testing. Based on the statistical information provided in the EPRI document and the visual inspection and as-found testing described above, the most probable cause of the relief valves' failure to meet ASME acceptance criteria is attributed to corrosion bonding and/or internal binding. This failure mode is a commonly identified issue with relief valves that are in a wetted service environment. EPRI reports that approximately 70% of safety valve malfunctions are attributed to "setpoint drift" associated certain failure modes of which corrosion binding and internal binding are included.
EXTENT OF CONDITION
It is well known and acknowledged in the industry that relief valves used in wet service environments can experience corrosion disc-to- seat bonding as well as internal binding. At JAF, the following relief valves are within the IST program and are vulnerable to the failure modes associated with having a water medium:
10SV-40 RHR Shutdown Cooling Relief Valve 14SV-20A & B CS Pump Discharge Safety Valve 1ORV-41A to D RHR Pump A Suction Relief Valve 235V-34 HPCI Booster Pump Suction Safety Valve 1ORV-43A & B RHR HX Tube Side Relief Valve 23SV-66 HPCI Booster Pump Recirc Safety Valve 1ORV-46A & B RHR HX Shell Side Relief Valve 46RV-112A to D EDG Jacket Water Cooler ESW Outlet Relief Valve 10SV-74A & B RHR HX Inlet Safety Valve 10SV-35A & B RHR Discharge Safety Valves Corrective actions are in place to evaluate the setpoint of these valves against the pipe Class pressure limit.
FAILED COMPONENT IDENTIFICATION
Manufacturer:� Crosby Valve and Gage Co.
Model No.:� JR-B NPRDS Manufacturer Code:� C710 NPRDS Component Code:� PSV FitzPatrick Component Id:
� 10SV-35A & B Manufacturer:� Crosby Valve and Gage Co.
Model No.:� 9811852E NPRDS Manufacturer Code:� C710 NPRDS Component Code:� PSV FitzPatrick Component Id:
� 14SV-20A 1. 10SV-35A, 10SV-35B, and 14SV-20A were replaced with pre-tested certified spares [complete].
2. Replace and test 14SV-20B. June 2010.
3. Engineering will re-assess the setpoint and tolerance for 10SV-35A, 10SV-35B, 14SV-20A, and 14SV-20B based on the pipe class pressure limits.
4. Incorporate relief valve test tolerance Level 1 and Level 2 acceptance criteria for owner specified limits in MP-059.07.
ASSESSMENT OF SAFETY CONSEQUENCES
10SV-35A and 10SV-35B, RHR "A" and "B" Loop Safety Valves 10SV-35A/B, "A" and "B" Loop RHR Safety Valves are designed to open to protect the RHR pump discharge piping when pressure in the piping exceeds the relief setpoint. Although these components have no specific accident mitigating or upset/abnormal functions, they support several RHR safety-related functions by protecting the RHR system loop piping integrity.
For the safety valves, the setpoint of 300 psig was chosen based on the design pressure of the RHR discharge piping. It is designed such that the pressure at the lowest point will not be greater than 110% of the design pressure during a thermally induced pressure increase. The JAF Piping Specification shows this piping class (302) to have a maximum operating pressure of 590 psig at 550 degrees Fahrenheit (F). The line designation table shows that the operating temperature for this section of piping, is 250 F. Therefore, 10SV-35A and 10SV-35B would have protected this piping from overpressure by lifting at 340 psig and 328 psig respectively.
14SV-20A, "A" CS Pump Discharge Safety Valve 14SV-20A, "A" Loop Core Spray Safety Valve is designed to open to protect the Core spray pump discharge piping when the pressure in the piping exceeds the relief set point. The maximum pressures that the system can potentially be exposed to are the pump shut off head and/or the pressure of the reactor. Although this component has no specific accident mitigating or upset/abnormal functions, it does support several Core Spray safety-related functions by protecting the Core Spray system loop piping integrity.
14SV-20A failed to lift within the required +/- 3% during setpoint testing. The actual lift occurred at 530 psig, which is 6% over the 500 psig setpoint. The JAF piping specification shows this piping class (302) to have a maximum operating pressure of 590 psig at 550 F.
The line designation table shows that the operating temperature for this section of piping is 213 F. Since the relief valve lifted at 530 psig, the CS piping would have been protected from overpressure.
Conclusion Based on the engineering review of the system design pressures and lift setpoints, the safety valves would have performed their safety function to protect their respective system from over-pressurization. Therefore, this condition did not constitute a loss of safety function.
SIMILAR EVENTS
A review of the corrective action program at JAF was performed to identify any similar events that involved the failure of safety/relief valves to meet IST acceptance criteria. The review looked at LER(s) and condition reports for the previous 3 years. The review found that the Emergency Diesel Generator Jacket Water Cooler Emergency Service Water Outlet Relief Valves (46RV-112A thru D) have also experienced IST setpoint test failures in which the apparent cause was corrosion bonding. Corrective actions are already in place to replace these valves with alternate relief devices (rupture discs) in JAF's next refueling outage (R19).
REFERENCES
1. Apparent Cause Evaluation Report: 10SV-35A/B IST Setpoint Failure (CR-JAF-2010-01382) 2. Apparent Cause Evaluation Report: 14SV-20A IST Setpoint Failure (CR-JAF-2010-01595) 3. JAF Technical Specifications 3.5.1, Emergency Core Cooling Systems — Operating 4. JAF Procedure: MP-059.07, Testing of Relief and/or Safety Valves (1ST) 5. ASME 0Mb Code-2003 Addenda to ASME OM Code-2001, Code for Operation and Maintenance of Nuclear Power Plants