arXiv:2609.17694
Efficiency-Resolved Recovery Dynamics of an Free-Running InGaAs/InP Single Photon Avalanche Detector Operated in Gated Mode
Bruno Santos Souza, Vitor Tavares, Filippo Ghiglieno, Celso Jorge Villas-Boas, Paulo Henrique Dias Ferreira
Abstract
Detection efficiency in gated-mode single-photon avalanche detectors recovers gradually after each gate edge, limiting attainable count rates, yet quantitative recovery models exist only for free-running operation. We extract the gated recovery time constant of a free-running InGaAs/InP detector at 1550 nm by sweeping the gate frequency from 100 kHz to 1 MHz and globally fitting count rates across four dead times, requiring neither a pulsed laser nor a time-to-digital converter. The recovery time decreases linearly with detection efficiency, from 301 to 161 ns between 10 % and 25 %, at 9.3 ns per percentage point. We further resolve a periodic count-rate modulation consistent with bias-dependent capacitive gate feedthrough.
AI Summary
Gated InGaAs/InP single-photon avalanche detectors can have a finite efficiency rise time after each gate edge. This paper asks how that transient depends on the selected detection efficiency and whether it can be inferred from count rates without pulsed illumination or time-tagging hardware. A commercial free-running detector is operated through its external gate input under continuous-wave illumination at 1550 nm. Gate frequency is swept from 100 kHz to 1 MHz at 50% duty cycle, for four nominal efficiencies between 10% and 25% and four dead times between 10 and 80 $\mu$s. Count-rate curves are globally fitted using an exponentially rising efficiency integrated over each gate, together with non-paralyzable dead time; an extended model adds discrete re-arming and a phenomenological sinusoidal modulation. The fitted recovery constant decreases from $300.9\pm3.3$ ns at 10% efficiency to $161.4\pm3.2$ ns at 25%, approximately linearly with slope $-9.28\pm0.28$ ns per percentage point. Adding the modulation changes these estimates little while improving fit criteria. Fitted modulation amplitudes span 1.56–3.59%, with periods in gate-frequency space of approximately 704–864 kHz; residual periodograms show corresponding structure. The efficiency settings are not independently calibrated, and excess bias is not measured. The frequency/dead-time grid makes the discrete re-arming correction identically inactive, leaving it untested. Count rates do not separate blind delay from recharge dynamics, the capacitive-feedthrough interpretation is not directly verified, and extended-model reduced chi-square values of 5.3–21.6 indicate remaining discrepancies.
Central claim may not hold: The paper interprets a zero-delay exponential fit as an intrinsic recovery constant even though it acknowledges that blind delay and recharge dynamics have not been separately identified. The linear efficiency dependence is established only against uncalibrated nominal detector settings, rather than independently measured detection efficiencies. Substantial residual model discrepancy remains after the Full Model fit, and inflating statistical uncertainties does not establish that the extracted recovery parameters are unbiased. The claimed exact saturation treatment retains a linearized per-gate detection probability, without testing its effect on the extracted parameters.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 8 pages, 6 figures, 2 tables