Interpretation of subcriticality measurements with strong spatial effects

This thesis presents a technique to identify and isolate modal or spatial effects in frequency-domain quantities used for subcritical reactivity measurements using the ²⁵²Cf-source-driven noise analysis technique. The ²⁵²Cf technique is based on measurements of the system's response to a stochastic Cf neutron source with three detectors. Detector 1 measures the source fissions, while 2 and 3 measure the system's response at different locations. If point kinetics conditions are valid, it has been shown that the reactivity of the system can be related to the frequency-independent ratio of the cross power spectral densities between detectors. For some geometries and detector configurations, the point kinetics approximation is not valid; spatial effects (i. e. modal contamination) are important and special interpretation is necessary. Removal of these spatial effects has been accomplished in the past by parallel neutronic calculations of the modal contamination factors. This approach has been successful in simple geometries.

The approach taken in this thesis is to functionally fit the measured CPSDs using a model that accounts for the modal contamination. The fundamental-mode component of the measured quantities is extracted from the fit, and a fundamental-mode ratio is calculated. All the current theories can be now applied. Subcritical reactivities can, in this way, be estimated using the existing proven theories even in cases where the geometry or the modal contamination would invalidate the point kinetics approximation.

To validate the model, data from experiments involving ²³⁵U solution tanks is analyzed. How spatial effects can be diagnosed and isolated is shown. Finally, reactivity calculations are made using a previous developed two point kinetic model. The results obtained agree with Monte Carlo simulation using the code KENO.