$\mu$ TCA DAQ System and Parallel Reading in CANDLES Experiment

The CAlcium fluoride for studies of Neutrino and Dark matters by Low Energy Spectrometer (CANDLES) experiment is a study of the neutrinoless double beta decay of
⁴⁸
Ca. As this is an extremely rare decay, background rejection is very important in its assessment; therefore, the β-decay of
²⁰⁸
Tl remaining as background in the detector is removed by tagging the preceding α-decay of
²¹²
Bi. This tagging technique requires a short dead time in the data acquisition (DAQ) system. To reduce this dead time and enhance background suppression, a microtelecommunications computing architecture DAQ (μTCA DAQ) system has been implemented in the CANDLES experiment. MicroTCA is a standard for connecting modules via point-to-point data links in the backplane. In CANDLES, data from the flash analog-to-digital converter (FADC) modules are read over the data links using the SpaceWire protocol. The FADC modules are equipped with eight event buffers, which act as derandomizers to detect sequential decays. Flexibility in the DAQ network is enhanced by equipping each μTCA DAQ crate with a GigabitEthernet-to-SpaceWire interface to enable the use of off-the-shelf computers. However, due to its long turnaround time, the interface has a high latency (about 100 μs), therefore, to reduce dead time, we developed a DAQ system with four “crate-parallel” reading threads in which the modules are read in parallel. Implementation of this system reduces the readout time by a factor of four-from 40 down to 10 ms. This performance improvement is expected to result in reduced dead time and enhanced background suppression. Furthermore, “event-parallel” reading process (in which events are read in parallel) is introduced to increase the total throughput in energy calibration: with two “event-parallel” reading processes, the data rate is doubled.