France's Benchmark Results against Experimental Data from SPERT-IV Statics

The SPERT IV experimental program carried out in the sixties in the nuclear research centre of Idaho Falls, United States, was one of the major experiments dedicated to the Reactivity Initiated Accident (RIA) for Material Testing Reactors (MTR). Reactivity insertions between 0.8 and 1.8$ have been tested. CEA is interested in the interpretation of such experiments to validate the modelling in the thermal hydraulics and neutronics fields. The SPERT IV core has been modelled with the French Monte Carlo code TRIPOLI4 in a heterogeneous description. The most recent nuclear data library JEFF3.1.1 has been used. All the static neutronics measurements have been studied. A dedicated software in C++ language has been developed for the modelling of the SPERT IV experimental configurations. To validate this tool, the reference loading of 25 fuel assemblies was studied with control rods at the experimental critical positions. The comparisons between the results of the calculations and the experimental measurements show a good agreement for most of the parameters except for the moderator temperature coefficient having a low reactivity effect. Indeed, the axial and radial power distributions are needed as an entry for the modelling. The reference configuration is the state of the core before and after the transient rod ejection. The corresponding axial position of the transient rod was determined from the transient rod efficiency curve that was calculated for this purpose. There is no experimental data for those axial and radial fluxes distributions. (author)