Skip to main content
Back to timeline
arXivSource publication:

T0 detector doubles as beam monitor: 32 SiPM channels regress 2D position, about 2.4 mm in experiment

Synopsis

The T0 detector of the HIRFL-CSR external-target experiment previously supplied only the event start time; its 32 SiPM over-threshold times and TOT values are reorganized into a geometry-preserving tensor, and a CNN regresses the incident coordinates directly. Residual standard deviations are about 1.11 mm and 1.04 mm for uniform planar simulation, about 2.36 mm and 1.34 mm for the laser-scanning geometry in simulation, and about 2.42 mm and 1.41 mm in experiment. The trained model reconstructed a two-dimensional beam spot from CEE zero-field beam data taken at the Institute of Modern Physics, Lanzhou.

Source-provided article image: CNN-based Position Reconstruction for the T0 Detector in the HIRFL-CSR External-target Experiment
Figure 1 · arXiv

Interpretation

Beyond the event start time, the T0 detector can also give the two-dimensional position of incident particles from the same multi-channel readout, letting it act as a high-rate beam-spot monitor. The dedicated beam-profile detector in CEE has a rate capability of only about 1 kHz, far below the designed beam event rate of up to 1 MHz, so it cannot reflect the beam-spot status in real time under normal data taking and has no event start-time function. On pulsed-laser scanning data with known hole-position labels, the experimental position resolutions are about 2.420 mm in x and 1.406 mm in y, close to the 2.364 mm and 1.337 mm from the simulation with the same scanning geometry.

The position information comes from the joint response pattern of the 32 SiPM channels rather than from a single channel amplitude or a two-end time difference. Earlier position reconstruction mostly relied on channel geometric centroids, two-end time differences, or empirical calibration functions, while boundary reflections and path losses in the scintillator, SiPM gain drift with temperature and bias, and the nonlinear coupling of time walk and amplitude effects make analytical models hard to keep accurate over the whole detector. Each channel over-threshold time is corrected by event-mean subtraction and each TOT by event-wise trimmed-mean normalization, then rearranged by short-axis channel order and the correspondence between the two end arrays, and three convolutional modules with fully connected regression layers output the two coordinates.

The density of finite scan positions is one of the main limits on the present experimental resolution. Uniform planar simulation has continuous, spatially dense labels and gives about 1.111 mm and 1.039 mm; switching to the discrete scan-hole positions used in the experiment worsens the simulated resolution to about 2.364 mm and 1.337 mm. The experimental residual distributions are broader than the uniform planar simulation and show a certain position offset, related to optical-path alignment, scan-hole position errors, channel-response nonuniformity, and electronics noise in the real system.

The trained model was applied to real beam data and reconstructed an event-by-event two-dimensional beam-spot distribution. Previously the T0 readout served only for the event start time, while beam-spot monitoring depended on a separate low-rate detector. On data collected during the CEE zero-field beam experiment at the Institute of Modern Physics, Lanzhou, in January 2026, the sample contains about 7854 events, and the mean and standard deviation of the reconstructed beam-spot position are reported in millimeters.

Perspective

The results cover uniform planar simulation, scanning simulation with the same geometry as the experiment, pulsed-laser scanning data with known hole labels, and beam-spot reconstruction on CEE zero-field beam data taken at Lanzhou in January 2026. They apply to the 32-channel SiPM readout of the T0 detector in the HIRFL-CSR external-target experiment, for researchers and engineers working on beam diagnostics and detector data reconstruction. Absolute beam-spot coordinates still require calibration against the detector installation coordinate system and the beam-line coordinate system.

The experimental resolution is close to the simulation with the same scanning geometry, indicating that the spatial sampling density of the finite scan positions is one important factor limiting the present resolution; laser-coupling stability, hole-position machining and installation errors, channel-response nonuniformity, and electronics-threshold differences may also contribute to the residual broadening. Absolute beam-spot coordinates are not yet calibrated, and the long-term stability of the model in beam-experiment data remains to be studied.