Search

As a part of the preparation of human building capabilities in Egypt in neutronic calculations for power reactors, a preliminary step was to perform calculations on benchmark problems related to neutronics calculations on lattice and core levels. Monte Carlo codes are capable of calculating integral parameters allowing all geometrical details of each individual fuel pin of each fuel assembly to be modeled. However, when more detailed information is required, it becomes much more difficult to get reliable results with respect to the associated statistical uncertainty within an acceptable computation time. Nonetheless, this is the requirement of current design calculations of nuclear reactor core. In order to assess the problems of Monte Carlo results for pin power distribution, this paper presents the results of the calculations of the C5G7 NEA benchmark problem using Monte Carlo code MCNP5. The present calculations used 50 & 90 million histories and 0.354% & 0.312% RMS statistical pin power percent error were achieved for the three-dimensional configurations. Results were compared with similar published results. Good agreement was found for the value of keff with the reference solutions. For the pin power distribution, discrepancy is large for UO2 assembly. The results for MOX assembly are much better. [ABSTRACT FROM AUTHOR]