The localization of a neutron source in an unknown homogeneous medium; the 3D case

Abstract

The possibility of localization of an unknown neutron source in a bulky homogeneous medium (box) was studied. Previous study showed that for the planar case, a neutron source could be localized with accuracy better than 7%, by using the count rates ratio of two 3He detectors, positioned on the opposite faces of the box. The error is larger close to the edge of the box. Positioning of polypropylene slabs of different thicknesses between the detectors and the box, in order to cancel the effect of the edge of the box, was found to improve the accuracy of the localization of the source in points close to the faces of the box. The influence of the deviation of the source from the plane connecting the two 3He detectors was studied. It was found that for detectors placed on the centers of the sample faces, and neutron sources positioned at a large distance from the central plane, the calculation of the vertical distance between the source and the box plane by the usual R( x)/ R(0) ratio with R(0) measured at the center of the box plane leads to quite large error, up to 15%. However, it was found that the error can be reduced by measuring R(0) of external source placed on the box face coplanar with the unknown source. In this case the accuracy does not depend on the distance from the central plane. Different ways to determine the plane of the unknown source were studied. It was found that it is impossible to determine it by looking for maximum count rate of one detector or the geometric mean of pair of detectors. However, it is possible to get better localization results of the source plane by using several pairs of detectors and finding a pair with maximal geometric mean. The experimental results were confirmed with MCNP simulations. It was found that using six detectors, three on each parallel faces of the box (460×420×200 mm 3), the location of the source can be found with an average error of 2.3% and a maximal error of 11%.