Digital frequency domain multiplexing readout electronics for the next generation of millimeter telescopes

TL;DR

This paper introduces a next-generation digital frequency domain multiplexing readout system for millimeter telescopes, significantly increasing the multiplexing factor to support large detector arrays while maintaining low noise and stability.

Contribution

The paper presents a new fMux readout system with a multiplexing factor of 64, enhanced bandwidth, active feedback, and digital synthesis, suitable for space-based applications.

Findings

Multiplexing factor increased to 64 channels per module.

Maintains low noise and detector stability at higher multiplexing.

Features include low power consumption and radiation-hard components.

Abstract

Frequency domain multiplexing (fMux) is an established technique for the readout of transition-edge sensor (TES) bolometers in millimeter-wavelength astrophysical instrumentation. In fMux, the signals from multiple detectors are read out on a single pair of wires reducing the total cryogenic thermal loading as well as the cold component complexity and cost of a system. The current digital fMux system, in use by POLARBEAR, EBEX, and the South Pole Telescope, is limited to a multiplexing factor of 16 by the dynamic range of the Superconducting Quantum Interference Device pre-amplifier and the total system bandwidth. Increased multiplexing is key for the next generation of large format TES cameras, such as SPT-3G and POLARBEAR2, which plan to have on the of order 15,000 detectors. Here, we present the next generation fMux readout, focusing on the warm electronics. In this system, the multiplexing factor increases to 64 channels per module (2 wires) while maintaining low noise levels and detector stability. This is achieved by increasing the system bandwidth, reducing the dynamic range requirements though active feedback, and digital synthesis of voltage biases with a novel polyphase filter algorithm. In addition, a version of the new fMux readout includes features such as low power consumption and radiation-hard components making it viable for future space-based millimeter telescopes such as the LiteBIRD satellite.