Reactivity Investigations Using Periodic Neutron Source in a Source in a Sub-Critical Assembly

Abstract

A method of reactivity measurement in the far subcritical range has been presented using a lumped-parameter model of a reactor whose neutron field is excited by a sinusoidally varying source of fast neutrons. Essentially, the method consists of determining the source transfer function for the system and then relating the parameters of the source transfer function to the subcritical reactivity. It appears that this method has several advantages over conventional methods in that the power level of the reactor does not have to be exactly steady, since an alternating component of the power is measured rather than the level itself. Furthermore, the, method lends itself to on-line measurements with as little disturbance as possible to an operating system. With the availability of digital computing facilities which today are becoming commonplace, the results can be made available immediately; on the other hand, it is feasible to record the raw data and process it later. A natural uranium-light water moderated subcritical assembly has been supplied with a square-wave input of 14 MeV neutrons produced by the SAMES type J accelerator in order to test the validity of this method. By varying the square-wave frequency and also Fourier analysing the resultant detector signal the source transfer function of the assembly has been measured as a function of frequency. The neutron signal was detected by a scintillation counter with a lithium glass scintillator and fed to a multichannel pulse-height analyser used in the time sequence storage mode. The experimental equipment was assembled from commercially available units in conjunction with some pulse circuitry which was developed using semiconductor devices. Calculations have also been performed using one- and two-group diffusion theory and transport theory to determine the source transfer function for a lumped-parameter model of the subcritical assembly as a function of negative reactivity and source frequency. The results are then discussed in terms of the theoretical relations derived and the physical phenomena taking place. Reasonable agreement between theory and experiment in the general shape of the transfer function was obtained, but the experimental results show that the transfer function is space and energy dependent. The results also indicate that the variation of the transfer function with changes in subcriticality is similar to that predicted theoretically. At frequencies of about 1000 Hz, it appears that the source transfer function method can be used to measure shutdown reactivities in the region of -50 dollars, provided that due allowance is made for the space and energy dependence of the transfer function. Also, results indicate that this method could be utilized to design a reactor criticality safety monitor. Investigations carried out to establish the validity of the method reveal that at angular frequencies in excess of  (equation see thesis) radians/sec, where v is the velocity of neutrons and L the diffusion length),neither diffusion or transport theory is satisfactory.