Measuring dark currents in multiple cryogenic SiPMs with sub-pA sensitivity using an automated IV multiplexer

TL;DR

The paper addresses the challenge of high-sensitivity dark-current measurements in large cryogenic SiPM arrays by introducing an automated, vacuum-compatible IV multiplexer (IV-MUX). The design enables up to 105 channels controlled by a single Arduino, using guard rings and high-isolation relays to achieve sub-$49$ fA resolution, while supporting both static IV and dynamic pulse-counting modes. Validation occurs through vacuum tests with a 16-channel SiPM array and LN tests with a 4-channel SiPM, showing excellent agreement between IV and dynamic measurements and revealing practical considerations like burst events influencing the effective DCR. The open-source IV-MUX reduces measurement time and hardware complexity for large-scale detector tests, facilitating rapid, automated characterization of SiPMs and similar low-current devices in cryogenic and vacuum environments.

Abstract

We present the design of an automated current-voltage (IV) multiplexer (MUX) that enables accurate measurement of the dark current in cryogenic silicon photomultipliers (SiPMs), achieving a sensitivity equivalent to detecting less than one avalanche per second. Dynamic pulse-counting measurements were used as a benchmark for reconstructing the dark current in static IV measurements. The IV-MUX features 15 channels on a single board and up to seven boards can be connected in parallel under the control of one Arduino microcontroller. To minimize leakage and enhance performance, the layout includes guard rings and high-isolation relays, enabling resolution of currents as small as 49 fA. The IV-MUX can be integrated into systems designed for IV or pulse-counting measurements, enabling seamless switching between IV and pulse-counting modes. Moreover, the IV-MUX is vacuum-compatible, validated by testing an SiPM array in a cryostat. This feature reduces the need for multiple feedthroughs when testing sensor arrays in vacuum. The design is open source and can be used to facilitate rapid and automated testing of SiPMs or similar low-current devices in one measurement cycle.

Figures (11)

• Figure 1: Left: Annotated IV-MUX in shielded enclosure for bench-top use. (A) 15 bias connectors, (B) sense line relays, (C) Arduino, (D) serial communication, (E) shift registers, (F) triaxial readout, (G) board interconnect, (H) bias line relays, (I) 12 V board power and bias power, (J) sense line input for mode B (described in text). Right: A Hamamatsu SiPM inside an aluminum enclosure, mounted on a custom PCB designed for routing signals in either mode A or mode B of the IV-MUX, which is described in Section \ref{['sec:2design']}. Each quadrant of the device is segmented, allowing four independent channels, which are labeled. The top of the enclosure, which includes an optical port and a demountable LED, is shown next to the SiPM. The SiPM is used for measurements in LN, which are presented in Section \ref{['sec:4airtest']}.
• Figure 2: (A) A simplified circuit diagram showing the IV-MUX switches and current paths on the MUX. Red traces are guarded by the nodes indicated with the dotted lines. A combined sense node configuration is shown for use with a single ammeter - isolating this interconnect would allow each board to be used with a separate ammeter. (B) A single SiPM connected in mode A. (C) Two SiPMs connected in mode B.
• Figure 3: Left: The IV-MUX prepared for vacuum use. The system uses one triaxial connection to the electrometer for current readout, and one 9-pin Subminiature C feedthrough for power, bias voltage and communication with the Arduino. In this configuration, it is equipped for measurement of up to 15 SiPMs. Right: A Hamamatsu SiPM mini-tile mounted atop a custom PCB configured to interface with both modes of the multiplexer, as explained in Section \ref{['sec:2design']}. Header pin jumpers are used to switch between the two modes. The 16-channel SiPM array was used for measurements within a cryostat presented in this section. The two aluminum lugs housing RTDs are in the top-left and top-right corners of the PCB.
• Figure 4: Dark IV measurements of 12 SiPMs using mode A (left) and mode B (right) of the IV-MUX. The standard deviation of the pre-breakdown region, which provides an estimate of the minimum signal resolution, was calculated and is shown in the legend. A constant value was added to each IV curve to align the pre-breakdown region for all SiPMs, which facilitates visual comparison. The uncertainties are a combination of statistical and instrumental uncertainties, added in quadrature.
• Figure 5: Left: A schematic of the experimental setup. The SiPM is inside of an aluminum enclosure that is outfitted with an LED and neutral density filters. The enclosure is placed within two nested dewars and submerged in LN. Right: A simplified electrical diagram of a single channel for IV and pulse-counting measurements. The high-side of the SiPM is used for pulse-counting measurements, and the low-side for IV measurements. The bias node of the IV-MUX is connected to the bias filter, which is wired in parallel to the SiPM cathode and the decoupling capacitor (which is connected to the amplifier). The current sense node of the IV-MUX is connected to the anode of the SiPM.