Fast Digitizing and Digital Signal Processing of Detector Signals

• Main Author: Hannaske, Roland
• Language: English
• Published: Forschungszentrum Dresden-Rossendorf 2010
• Subjects: info:eu-repo/classification/ddc/004; ddc:004
• Online Access: http://nbn-resolving.de/urn:nbn:de:bsz:d120-qucosa-27888; https://hzdr.qucosa.de/id/qucosa%3A21615; https://hzdr.qucosa.de/api/qucosa%3A21615/attachment/ATT-0/

A fast-digitizer data acquisition system recently installed at the neutron time-of-flight experiment nELBE, which is located at the superconducting electron accelerator ELBE of Forschungszentrum Dresden-Rossendorf, is tested with two different detector types. Preamplifier signals from a high-purity germanium detector are digitized, stored and finally processed. For a precise determination of the energy of the detected radiation, the moving-window deconvolution algorithm is used to compensate the ballistic deficit and different shaping algorithms are applied. The energy resolution is determined in an experiment with γ-rays from a 22Na source and is compared to the energy resolution achieved with analogously processed signals. On the other hand, signals from the photomultipliers of barium fluoride and plastic scintillation detectors are digitized. These signals have risetimes of a few nanoseconds only. The moment of interaction of the radiation with the detector is determined by methods of digital signal processing. Therefore, different timing algorithms are implemented and tested with data from an experiment at nELBE. The time resolutions achieved with these algorithms are compared to each other as well as to reference values coming from analog signal processing. In addition to these experiments, some properties of the digitizing hardware are measured and a program for the analysis of stored, digitized data is developed. The analysis of the signals shows that the energy resolution achieved with the 10-bit digitizer system used here is not competitive to a 14-bit peak-sensing ADC, although the ballistic deficit can be fully corrected. However, digital methods give better result in sub-ns timing than analog signal processing.