Study of the Front End Electronics Contribution to the Dead Time in He3 Proportional Counters

This paper will describe a study of the contribution of front end electronics to dead time in 3 He thermal neutron detectors. The well known source of dead time in all detectors is the duration of a pulse created by the convolution of the detector pulse with the preamplifier/shaper impulse response. Some detectors have non-uniform pulse shape and duration. In 3 He proportional counters, the duration and non-uniformity of detector pulse shapes is the dominant contributor in that convolution. Therefore the contribution of the electronics is masked and often lumped into the overall dead time analysis. For example this is the case in time interval analysis, which is a very common method for determining dead time. The contribution of the electronics, however, is not negligible. It increases the dead time by stretching the pulse further. Here we present a study of how the detector pulse properties are manifested in the more widely used front end devices in neutron detectors/coincidence counters, namely Amptek A111 and PDT-10 shaper/discriminators. In order to study the electronics contribution, we developed an experimental setup that allowed us to inject charge pulses with controllable amplitude and relative phase at the input of the detector electronics and thus measure pulse pair resolving time for different settings of amplitudes by the observing the discriminator output. We present DT measurement results for different modifications of Amptek A111 and PDT-110A shaper/discriminators at different gain settings and signal levels. The effect of shaping circuitry saturation on DT when operated with high signal level will be discussed also. Recommendations regarding optimal tube and shaper gain settings will be provided. This study also yielded evidence for non-linear shaping and saturation in A111 and PDT, along with double pulsing of the logic output for certain detector pulse shapes, which could also be observed in time interval analysis. Double pulsing has implications for most detector applications. Techniques to mitigate this, such as blanking, increase dead time and unrecognized pileups. They also cause non-linearity and time variance of the system. The presented data shows how dead time varies with pulse amplitude in the studied devices, and what effects and non-idealities it adds to the detector chai