NUREG/IA-0176 Post-Test Analysis of P5 Experiment in PANDA Facility With TRAC-BF1 Code

As a part of the TEPSS project, within the last fourth Framework Programme of the European Union R&D activities, an investigation focused on the residual heat removal passive safety systems in the future boiling water reactors was performed using PANDA facility and the codes RELAP5, MELCOR and TRAC-BF l. The main purpose was to ensure the capacity of these passive safety systems under a wide range of different conditions and hypothesis on the availability of them as well as to assess the quality of some of the most commonly used codes to predict the containment response within a LOCA. The plant design chosen as reference was the ESBWR, which is a special version of the SBWR adapted to reach licensing conditions in European countries. The PANDA facility, located in the Paul Scherrer Institute (Switzerland), has been extensively used in the past for testing the passive condenser performance for the SBWR. According to the reference plant, PANDA facility was appropriately modified. A set of eight tests was executed covering different conditions in the containment for the LOCA scenario, such as asymmetric conditions for the gas distribution in the drywell as well as for the passive condensers availability and the effect of trapped air in the drywell among others. The assessment of codes was rather challenging, since RELAP5 and TRAC-BF1 are thermal-hydraulic codes usually employed in transients within the primary circuit in a current plant. Thus, they have a well proved response in the conditions were strong changes of the momentum are expected, like normally occur in forced convection situations associated with valves and pumps. However, in containment buildings equipped with passive safety systems, the expected phenomena are related with changes in the density, that is natural convection. In addition, some other phenomena like condensation in the presence of non-condensable gases are also characteristic in the scenario. For that reason, there were also MELCOR predictions, since this code incorporate containment modelling, although it has no special models for the passive condensers in the containment. The general response of all the codes was rather acceptable, being all of them able to predict well the overall behaviour of PANDA concerning pressure, mass flow rates and pool boil-off. Nevertheless, some limitations were showed regarding 3D-effects such as in-vessel mixing and recirculation. This report describes the calculations performed with TRAC-BF1 code for P5 test.