Monte Carlo calculations of the REBUS critical experiment for validation of burn-up credit

The REBUS experiment is a valuable benchmark for the validation of Monte Carlo criticality codes within the context of burn-up credit. It investigates the difference in reactivity worth of unirradiated fuel rods and of fuel rods irradiated in a PWR. This paper presents results of criticality calculations for this experiment with the Monte Carlo codes MCNP and SCALE/KENO. The results show that even the comparatively small reactivity loss of about 2000 pcm measured in this experiment can be calculated within an accuracy of 6 per cent. A detailed comparison of measured and calculated core characteristics (fission rate and flux distributions) shows good agreement, which indicates an adequately detailed model geometry for criticality calculations. The slight underestimation (500 pcm) of absolute k(eff) values is not out of the typical range for criticality benchmark experiments