The Demonstrated Function of PLR Pump Mechanical Seal Under SBO Conditions

Abstract

Generic Safety Issue 23 (GSI-23) by the U.S. Nuclear Regulatory Commission (NRC) was identified in 1980 as a result of staff concerns about reactor coolant pump (RCP) seal failure during Station Blackout (SBO), that is, seal degradation leading to a significant loss of reactor coolant in pressurized-water reactor (PWR) plants. Resolutions of GSI-23 have been considered at PWR plants. In 2000, NRC decided to close GSI-23 and issue Regulatory Issue Summary00–002 (RIS 00–002)[1], based on considerations such as; for example, the improvement of RCP seal performance[2], and the reduction of the risk of RCP seal failure in certain plants by the addition of alternate power sources. After the closure of GSI-23, some licensees were planning to make other associated improvements under their individual plant program. In Japan, the RCP seal was showed that leakage rate was low under SBO testing conditions[3] in licensing safety reviews conducted according to new nuclear regulatory standards after the nuclear accident at the Fukushima Daiichi Nuclear Power Plant. On the other hand, a boiling-water reactor (BWR) is not included in GSI-23 because operating experience indicates that seal failures in BWRs result in smaller leak rates than seal failures in PWRs and, BWRs have the reactor coolant injection capability under SBO conditions, such as the reactor core isolation cooling (RCIC) and the high-pressure coolant injection (HPCI) system. In addition, for the particular BWR-2 type plants that do not have emergency makeup systems, the pump mechanical seal was tested under SBO conditions and successfully showed minimal leakage[4]. However, for the BWR-5 type plants which have the reactor coolant injection capability, such as the RCIC and HPCI systems, the pump seal had not been tested. In Japan, after the nuclear accident at the Fukushima Daiichi Nuclear Power Plant, licensing safety reviews of BWRs and PWRs are being conducted according to new nuclear regulatory standards. We took this opportunity to test the leak rate from Primary Loop Recirculation (PLR) pump mechanical seals under SBO condition. The peak of leak rate was approximately 0.6ton/h (2.6gpm) during the 24 hours of SBO testing condition. Despite damage of O-rings in the mechanical seal by heated water which were ovserbed at post-inspection test, a very low leak rate was realized because the leakage path after passing through the damaged parts of the O-rings was limited by the other restricting pathway in the mechanical seal. This seal leakage was very low, compared with the reactor coolant makeup capability of the RCIC system and the reactor coolant release capability from main steam safety relief valve (SRV). Therefore, we reconfirmed that the result of this leak rate does not affect the safety evaluation for the reactor. It is shown in this paper that the leak rate from PLR pump mechanical seals is low under SBO condition by our demonstration test.