Abstract
A <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sup> LiF:ZnS(Ag)-based cold neutron detector with wavelength shifting (WLS) fibers and Silicon photomultiplier (SiPM) photodetector was developed at the NIST Center for Neutron Research. For neutron scattering applications at the NCNR, detector false positives severely diminish the quality of very faint neutron scatter patterns. Thermal noise generated by the SiPM significantly increases the likelihood of false positives by the detector/discriminator. This article describes and evaluates a digital real-time algorithm implemented on a field programmable gate array (FPGA) which quickly differentiates SiPM thermal noise and noise pulse pile-up from neutron signals. The algorithm reduces deadtime spent on examining noise pulses as well as reduces the number of false positives.
Highlights
S ILICON photomultipliers (SiPMs) are devices which greatly amplify a very low light signal [1]–[6]
With gains above 1E+ 6 electrons per photon, detection of single photons is possible. This is similar in concept to a photomultiplier tube, but Silicon photomultiplier (SiPM) have some advantages over a photomultiplier tube, as well as some disadvantages
The CANDOR_DAQ (Fig. 2) is an in-house analog-to-digital converter/field programmable gate array (ADC/FPGA) solution. It uses a Texas Instruments AFE5801 VGA/ADC sampling at 50 mega-samples per second (MSPS) and 12-bit resolution and a Xilinx Artix-7 FPGA for signal processing
Summary
S ILICON photomultipliers (SiPMs) are devices which greatly amplify a very low light signal [1]–[6]. There are signal nonlinearities such as thermal noise, crosstalk, and afterpulsing which are common [7]–[10]. The National Institute of Standards and Technology (NIST) Center for Neutron Research has developed a 6LiF:ZnS(Ag) scintillating neutron detector with wavelength shifting (WLS) fibers [11], [12] for routing light to a SiPM. Similar cold/thermal detector schemes, utilizing 6LiF:ZnS(Ag) and WLS fibers, have been developed at other institutions [21]–[24]. SiPM thermal noise, crosstalk, and afterpulsing confound traditional PSD techniques, making it challenging to meet the performance demands for neuron scattering applications which include: very high gamma rejection (fewer than one false detection per 1E+ 7 incident gamma rays), high neutron detection efficiency (near 90% detection of cold neutrons), and minimal deadtime. Some proposed PSD algorithms which record data and retroactively process a waveform using wavelet transforms and/or artificial neural networks are not considered due to the relatively large amount of computing resources needed [49], [51], [57], [59], [61], [62]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
