RELAP5-3D currently calculates two-phase pump degradation using the Aerojet Nuclear Corporation (ANC) model. This is an empirical model that relates two-phase pump performance to single-phase pump performance using a set of two-phase degradation multipliers, which are only a function of void fraction. The purpose of the present work was to assess the two-phase pump degradation model in RELAP5-3D and various sets of user-supplied two-phase degradation multipliers by modeling a full-scale, two-phase pump test facility and comparing the simulated results to experimental data. Tests conducted by Ontario Hydro Technologies (OHT) using a full-size CANDU reactor primary heat transport pump were used for this assessment. Presently, this work represents the only RELAP5-3D analysis of these tests that has been performed.

The experimental data from the OHT tests and results of this assessment both indicate that there is a pressure effect, in addition to void fraction, that cannot be neglected by safety analysis codes when predicting two-phase pump performance. The RELAP5-3D results showed that the widely used Semiscale two-phase head degradation multipliers did a poor job of predicting the experimental data and utilizing pressure-specific two-phase head degradation multipliers developed by OHT significantly improved code-to-data agreement. These results identify both the inaccuracies of using the Semiscale two-phase degradation multipliers and a weakness in the present formulation of the ANC model. As a result of this work, the Idaho National Laboratory recognized the need to include a pressure dependence in the RELAP5-3D calculation of two-phase pump performance, and this capability will be available in the next release of the code.