Parallel data readout for background rejection in CANDLES experiment

TL;DR

The paper presents a new parallel data readout system for the CANDLES experiment that significantly reduces dead-time, enabling more effective background rejection in the search for neutrino-less double beta decay.

Contribution

A novel DAQ system with parallel processing and multiple buffers that reduces read-time by fourfold and nearly eliminates data loss, improving background suppression capabilities.

Findings

Read-time reduced from 40ms to 10ms

Data loss minimized at 20 cps trigger rate

Enhanced background rejection potential

Abstract

CANDLES experiment in Kamioka Underground Observatory aims to obtain the neutrino-less double beta decay ($0\nu\beta\beta$) from $^{48}$Ca. This measurement is a big challenge due to the extremely rare decay rate of $^{48}$Ca. Thus, in order to obtain 0\nu\beta\beta, it is needed to reduce background as much as possible. Series of alpha and beta decays originated from radioactive impurities can remain as background in the energy region of $0\nu\beta\beta$. Because they are sequential decays, we can remove them by tagging preceding and following events. This tagging method requires minimized dead-time of DAQ system. A new DAQ system was introduced in CANDLES with new FADC (Flash Analog-to-Digital Converter) modules using 8 event buffers and SpaceWire-to-GigabitEthernet network for data readout. To reduce the dead-time, we developed our DAQ system with 4 parallel reading processes. As a result, the read-time is reduced by 4 times: 40msec down to 10msec, which is in equivalent to a half of previous DAQ's read-time. With reduced read-time accompanied by multiple event buffers, the new DAQ system is realized with efficiency is very close to 100$\%$ (no event lost at 20cps, which is CANDLES trigger rate, after 63 hours of data taking). With improved performance, it is expected to achieve higher background suppression for CANDLES experiment.