05000271/LER-2001-002
| Event date: | |
|---|---|
| Report date: | |
| 2712001002R01 - NRC Website | |
FACILITY NAME (1) DOCKET SEOUENTIA L NUMBER -
Description of Event
On 05/02/01, with the plant shutdown for refueling, it was determined during IOC-FRS°, Appendix J, Type C, primary containment leak rate testing that the leakage rate of both (in series) HPCI (EIIS=BJ) steam turbine exhaust nozzle check valves exceeded the Technical Specification (TS)'allowable leak rate limits. This event is reportable under 10CFR50.73(a)(2)(1)(B) as a condition prohibited by the plants TS and 10C1R50.73(a)(2)(v) as a condition that could have prevented the fulfillment of the safety function needed to mitigate the consequences of an accident.
The leakage of both HPCI steam turbine exhaust nozzle check valves, caused the combined local leak rate test (Type B and C tests) acceptance criterion of
Background
In 1999, VY installed two new nozzle check valves in the HPCI steam turbine 20" exhaust line. The nozzle check valves are similar and utilize a disc floating on springs and radial guides to check flow and have flange connections on each end for ease of installation/removal. These valves replaced two swing-check valves that were original plant equipment. They were satisfactorily leak rate tested after installation and not tested again until this refueling outage.
Both valves were designed to meet the specificationi of the application in the VY HPCI system. Those specifications took into account the need for HPCI to automatically start on demand and to permit operation under all design flow conditions while not creating excessive backpressure on the steam turbine. The valves are in series in the HPCI turbine exhaust line and the downstream valve will see a slightly lower operating pressure. Additionally one valve is physically located horizontally and the other vertically due to the existing piping configuration. The horizontal valve (V23-3) has stronger springs because the disc weight is being supported by the diffuser and radial guides and needs to be stronger to overcome the associated friction.
The springs in the vertical valve (V23-4) are designed with a lesser spring force because the weight of the disc aids in closing/seating the disc.
Cause of Event
The cause of the failure is concluded to be the design/sizing of the valve springs. The springs were specified to provide minimal spring force, to keep the HPCI turbine backpressure as low as possible. A contributing cause was believed to be the deformation of the radial guides. The spring design may have contributed to the deformation of the radial guides.
Analysis of Event
The HPCI system consists of a steam turbine assembly driving a constant-flow pump. Steam for the HPCI turbine is provided from the reactor and exhaust steam from the turbine is discharged to the suppression pool. The HPCI system is provided to assure that the reactor core is adequately cooled in the event of a small break in the nuclear system and loss of coolant which does not result in rapid depressurization of the reactor vessel, simultaneous with a loss of normal auxiliary power. There is no indication that HPCI system performance was negatively impacted during the previous operating cycle by any problem(s) associated with these nozzle check valves. Periodic HPCI system surveillance testing was satisfactorily performed during this period.
The primary containment consists of a drywell, which encloses the reactor vessel and recirculation system, a pressure suppression chamber which stores a large volume of water, a connecting vent system between the drywell and the suppression chamber, isolation valves and other associated components. The primary containment is capable of withstanding peak pressure which is postulated from a loss of coolant accident, to limit the release of fission products to the plant environs, so that off-site doses would be well below limits.
The leakage from the primary containment would be from the Torus air space via the HPCI vacuum breaker system (2" piping and two 3" in-series nozzle check valves) into the HPCI turbine steam exhaust systein (20" piping and two in-series nozzle check valves) and exceeded the leakage assumed in the current analysis. The leakage would be to the HPCI turbine, where it can be presumed to leak into the HPCI room (enclosed within secondary containment) through the turbine shaft seals. Any leakage into secondary containment through the turbine shaft seals would be captured, filtered and processed through the Standby Gas Treatment System (SBGT), thus limiting any potential off-site dose consequences.
The design basis LOCA for VY, as detailed in Section 14.6 of the FSAR, predicts the release of noble gases and iodine based upon conservatively estimated fuel failure and fuel gap inventories. Credit was taken for iodine removal in the drywell and torus, and by the standby gas treatment system. Conservative meteorology was used to determine off-site doses at a location where the terrain height is equal to the height of the primary vent stack. The resulting doses were in the millirem range for both whole body and thyroid. These doses are 3-4 orders of magnitude below the 10CFR100 criteria and any additional leakage through the HPCI turbine steam exhaust valves would not have resulted in an unacceptable risk to the public.
It has been determined that this condition would not adversely impact the plant's containment Level-2 Risk Assessment.
Specifically, this failure of primary containment integrity would not contribute to the determination of the Large Early Release Frequency (LERF) due to the small size of the air-space release flow path 2" diameter piping) and the fact any release via the main exhaust line discharge would be "scrubbed" by the suppression pool, reducing the postulated release magnitude by two release categories (i.e., from High to Low).
Corrective Actions
1. The valve springs were replaced with stronger springs. The HPCI system has been full flow tested and turbine backpressure is acceptable with the stronger springs. This action is complete.
2. The radial guides were replaced with stronger and thicker Inconel radial guides. This should preclude the potential for buckling or deformation of the radial guides. This action is complete.
3. The seat and disc for V23-3 was lapped to repair some minor seat damage. This action is complete.
4. Both valves were satisfactorily leak rate tested after repairs/modifications were complete.
5. Complete replacement of the check valves with design enhancements to improve long-term valve performance and reliability. This corrective action is scheduled to occur during RFO-23.
Additional Information
In the past 5 years, similar events have been reported as:
Ability of a System to Mitigate the Consequences of an Accident"