06/01613 Analysis of NEA C5MOX benchmark with computer code ‘COHINT’ based on interface current approach: Mohanakrishnan, P. Progress in Nuclear Energy, 2004, 45, (2–4), 125–131

Abstract

The 3-D extension C5G7 MOX benchmark, which is a follow-up benchmark of the C5G7 MOX on deterministic 2-D/3-D MOX fuel assembly transport calculations without spatial homogenization, aims at providing a more challenging test of the abilities of currently available 3-D methods to handle the spatial heterogeneities of reactor core. The geometry configuration was modified with some significant changes including control rod insertion in order to check real 3-D effects with the previous benchmark. In this work, THREEDANT code within DANTSYS code package, which solves the multi-group discrete ordinates form of the Boltzmann transport equation in x–y–z and r–z–theta geometries, was employed to perform the benchmark calculations with the spatial approximation to a square shape of the circular fuel pin and the angular approximation to S8. The calculation results from THREEDANT indicate that the power decrements due to control rod insertion are larger than those in the reference solution and are also more sensitive. In addition, a sensitivity study on mesh refinement was performed to produce a higher quality solution. In the results, it is found that the 4 × 4 grid per pin cell in the THREEDANT calculation is sufficiently refined so that very little benefit is obtained by increasing the mesh size.