OECD/UACSA criticality benchmark study to account for fabrication tolerances (Phase II), CRISTAL - SCALE - MCNP evaluations - 5-07

The evaluation of the calculated neutron multiplication factor K_eff of a fuel system has to account for: -) the bias due to both the biases on the nuclear data applied and the algorithmic and numerical weakness of the criticality code package used, -) the uncertainty in K_eff due to the uncertainties in the nuclear data applied, and -) the uncertainty in K_eff due to the uncertainties (manufacturing tolerances and/or variations) in the parameters describing the fuel system of interest. An OECD Expert Group was created in 2009 to address more particularly the analysis of the uncertainty in K_eff due to the manufacturing tolerances of the design parameters of a fuel assembly configuration derived from typical wet PWR fuel storage rack design and PWR fuel basket design for transport/storage casks. The UACSA (Uncertainty Analysis for Criticality Safety Assessments) working group gathers skills from both the industry and research laboratories. As an option and in order to benchmark the criticality code package used against real experiments some ICSBEP (International Criticality Safety Benchmark Evaluation Program) cases are also suggested. Even if the French criticality code package CRISTAL V1 is the basic code and methodology tool for AREVA Front End Business Group comparisons with other well known codes such as SCALE6 and MCNP4C2 were performed on those ICSBEP benchmarks and on the bounding cases of the OECD/UACSA benchmarks. Those later comparisons with a good agreement observed from the CRISTAL-SCALE-MCNP code packages give confidence for the criticality safety evaluations AREVA Front End Business Group have to perform either for its fuel manufacturing plants and fresh fuel shipping casks or for its abroad increasing criticality safety evaluations. The paper will focus on the full CRISTAL V1 parametric studies as requested by the UACSA working group as well as on some proposed ICSBEP critical experiments comparisons. As a conclusion of this paper it is observed that the three criticality code packages are well suited to assess criticality safety concerns. (authors)