Determination of large negative reactivity by integral versions of various experimental methods.

Abstract

The large negative reactivity is measured in Semi-Homogeneous Experimental facility (SHE). Experimental methods are Sjöstrand's pulsed neutron, source multiplication and rod drop methods beside revised King-Simmons' pulsed neutron methods. Neutron detectors are placed at various points in the core region for multi-points measurement. Usual one-point reactor model analysis resulted in the reactivity values, strongly dependent on the detector position with the increase of subcriticality. In addition, disagreements between the used experimental methods are also pointed out. In order to overcome these difficulties due to the spatial higher harmonics and the kinetic distortion in the neutron flux distribution, an integral version analysis is applied, in which use is made of multi-points reactor model. In the analysis, space integration of the neutron counts obtained throughout the core region is made with weights of the adjoint function of fast neutrons, calculated using the two- or three-dimensional diffusion code. The negative reactivity values determined by the integral version analysis agreed well with each other within the uncertainty of ∼5% in the reactivity range down to ∼50 dollars. It is concluded that all the experimental methods are adequate for precise determination of the large negative reactivity of reactor provided that the integral version analysis is utilized or that correction is made for the change of the neutron generation time using precise calculation.