Comparison of MCNP and WIMS-AECL/RFSP calculations against critical heavy water experiments in ZED-2 with CANFLEX-LVRF and CANFLEX-LEU fuels

This paper summarizes calculations of MCNP5 and WIMS-AECL/RFSP compared against measurements in coolant void substitution experiments in the ZED-2 critical facility with CANFLEX®-LEU/RU (Low Enriched Uranium, Recovered Uranium) reference fuels and CANFLEX-LVRF (Low Void Reactivity Fuel) test fuel, and H2O/air coolants. Both codes are tested for the prediction of the change in reactivity with complete voiding of all fuel channels, and that for a checkerboard voiding pattern. Understanding these phenomena is important for the ACR-1000® reactor. Comparisons are also made for the prediction of the axial and radial neutron flux distributions, as measured by copper foil activation. The experimental data for these comparisons were obtained from critical mixed lattice / substitution experiments in AECL's ZED-2 critical facility using CANFLEX-LEU/RU and CANFLEX-LVRF fuel in a 24-cm square lattice pitch at 25°C. Substitution analyses were performed to isolate the properties (buckling, bare critical lattice dimensions) of the CANFLEX-LVRF fuel. This data was then used to further test the lattice physics codes. These comparisons establish biases/uncertainties and errors in the calculation of keff, coolant void reactivity, checkerboard coolant void reactivity, and flux distributions. Results show small to modest biases in void reactivity and very good agreement for flux distributions. The importance of boundary conditions and the modeling of unmoderated fuel in the critical experiments are demonstrated. This comparison study provides data that supports code validation and gives good confidence in the reactor physics tools used in the design and safety analysis of the ACR-1000 reactor.