A Comparison of Some Analytical Models with Experiment for the University of Maryland Reactor

Abstract

It is often desirable in the operation of a research reactor to make adjustments in the nuclear core configuration. These adjustments may occur for a variety of reasons, such as the rearrangement of fuel to perform a particular experiment. It is beneficial to the reactor operator and experimenter to have an adequate analytical model with which to predict the changes in nuclear characteristics which occur with core rearrangement. Several analytical models have been investigated and compared with experimental results for the semipermanent, or normal, core configuration for the University of Maryland Reactor. These models were selected because, while somewhat time consuming with respect to the use of computers, the computer time utilized is much less than needed by more complex methods. At the same time, the methods used tend to minimize the large inherent error associated with simple hand calculations . The methods used consist of a two-dimensional few group diffusion theory coupled with several cross section models from which macroscopic cross sections were obtained. The cross section models used for the above thermal energy groups were the volume integrated P-1 method and the Fourier transform B-1 method. Thermal energy group cross sections we reobtained using the Wigner-Wilkins model and the Maxwell-Boltzmann model. The volume integrated P-1 model and the Wigner-Wilkins model coupled with the two-dimensional group diffusion method were found to give the best agreement with experiment for the semi-permanent core configuration. This model was then tested over a range of experiments. The conclusion of this analysis was that the model was capable of predicting, with reasonable accuracy, the changes in core reactivity with core rearrangement.