| Pile oscillators have been used for many years to cause small perturbations in a nuclear reactor and study some properties associated with the reactor itself and with the sample being oscillated. A sample oscillating in and out of a reactor core can be used to determine its reactivity worth through inverse kinetics (open-loop technique as described in Imel, et al., and it has been shown that cross-sections can be also calculated from the reaction rates with low uncertainties using perturbation theory techniques. Thermal cross-sections have been the object of study for many years and are well verified, both experimentally and through modeling. However, fast fission cross-sections for the next generation of advanced reactors have much higher uncertainties, with little experimental data to verify the values. Decreasing these uncertainties can help reduce future design costs and provide larger margins of safety. There has been interest in recent years to construct a large fast neutron sub-critical assembly to provide a platform for conducting measurements of various parameters including cross-sections. This assembly would be driven by a relatively modest neutron generator (e.g., D-T); the impetus is the belief that this could provide valuable data without having to construct an actual fast critical reactor with its substantial costs. However, most (if not all) of the perturbation/oscillation measurements in the past have been performed in a critical system. We are trying to answer the following questions in this study: 1) can meaningful measurements be conducted at sub-critical, 2) if so down to what level of sub-criticality, and 3) down to what level of source strength? In this dissertation we describe our
xi experiments in our AGN-201 reactor at critical and sub-critical (to about k=0.96) and our experiments in our Sub-Critical Assembly (maximum of around k=0.90).Key Words: Oscillations, reactivity, inverse kinetics, perturbation theory, subcritical, MCNP, ANSIN |