Simulation of Deadtime Behavior and the Effect on Performance and Design of Passive Neutron Multiplicity Counters

Neutron time-correlation measurements performed by thermal and epi-thermal neutron well counters and associated Multiplicity Shift Register (MSR) electronics are extensively used for quantifying spontaneously fissile materials such as Pu. Passive Neutron Multiplicity Counting (PNMC) is used to record the multiplicity histogram of detected neutron events falling within specified coincidence gates defined relative to the incoming pulse train. The Singles (Totals), Doubles (Reals) and Triples count rates may be calculated from the histograms. At high event rates, both instantaneous and sustained, deadtime losses perturb these rates. Without correction these perturbations are a major source of inaccuracy in both the measured count rates and assay values derived from them. Under certain operational conditions, for example, high (, n) emission rates, the deadtime corrections themselves may become the limiting factor in the assay result. This paper presents the results of a Monte Carlo study of deadtime behavior over a range of operational conditions of a nominal multiplicity counter. Results of the simulations demonstrate the change in the observed MSR data when the ratio of correlated to non-correlated neutron events within the pulse train was varied. In addition, deadtime behaviour was investigated for a range of counter deadtime parameters, characteristic of varying levels of counter performance. Deviations from conventional deadtime correction treatments are observed.