The development of planar fabrication techniques for superconducting transition-edge sensors has brought large-scale arrays of 1000 pixels or more to the realm of practicality. This raises the problem of reading out a large number of sensors with a tractable number of connections. A possible solution is frequency-domain multiplexing. I summarize basic principles, present various circuit topologies, and discuss design trade-offs, noise performance, cross-talk and dynamic range. The design of a practical device and its readout system is described with a discussion of fabrication issues, practical limits and future prospects.
The development of planar fabrication techniques for superconducting transition-edge sensors has brought large-scale arrays of 1000 pixels or more to the realm of practicality. This raises the problem of reading out a large number of sensors with a tractable number of connections. A possible solution is frequency-domain multiplexing. I summarize basic principles, present various circuit topologies, and discuss design trade-offs, noise performance, cross-talk and dynamic range. The design of a practical device and its readout system is described with a discussion of fabrication issues, practical limits and future prospects.
We describe the development of a frequency-domain multiplexer (MUX) to read out arrays of superconducting transition-edge sensors (TES). Fabrication of large-format arrays of these sensors is becoming practical; however, reading out each sensor in the array is a major instrumental challenge that is possibly solved by frequency-domain multiplexing. Each sensor is AC biased at a different frequency, ranging from 380 kHz to 1 MHz. The sensor signal amplitude-modulates its respective AC bias frequency. An LC filter associated with each sensor suppresses Johnson noise from the other sensors. The signals are combined at a current summing node and measured by a single superconducting quantum interference device (SQUID). The individual signals from each sensor are then lock-in detected by room temperature electronics. Test chips with fully lithographed LC filters for up to 32 channels have been designed and fabricated. The capacitance and inductance values have been measured and are close to the design goals. We discuss the basic principles of frequency-domain multiplexing, the design and testing of the test chips, and the implementation of a practical system.
We describe the development of a frequency-domain multiplexer (MUX) to read out arrays of superconducting transition-edge sensors (TES). Fabrication of large-format arrays of these sensors is becoming practical; however, reading out each sensor in the array is a major instrumental challenge. Frequency-domain multiplexing can greatly simplify the instrumentation of large arrays by reducing the number of SQUID's (superconducting quantum interference devices) and wires to the low temperature stages. Each sensor is AC biased at a different frequency, ranging from 380 kHz to 1 MHz. Each sensor signal amplitude-modulates its respective AC bias frequency. An LC filter associated with each sensor suppresses Johnson noise from the other sensors. The signals are combined at a current summing node and measured by a single SQUID. The individual signals from each sensor are then lock-in detected by room temperature electronics. Test chips with fully lithographed LC filters for up to 32 channels have been designed and fabricated. The capacitance and inductance values have been measured and are close to the design goals. We discuss the basic principles of frequency-domain multiplexing, the design and testing of the test chips, and the implementation of a practical system.
A technological milestone for experiments employing transition edge sensor bolometers operating at sub-Kelvin temperature is the deployment of detector arrays with 100s-1000s of bolometers. One key technology for such arrays is readout multiplexing: the ability to read out many sensors simultaneously on the same set of wires. This paper describes a frequency-domain multiplexed readout system which has been developed for and deployed on the APEX-SZ and South Pole Telescope millimeter wavelength receivers. In this system, the detector array is divided into modules of seven detectors, and each bolometer within the module is biased with a unique ∼MHz sinusoidal carrier such that the individual bolometer signals are well separated in frequency space. The currents from all bolometers in a module are summed together and pre-amplified with superconducting quantum interference devices operating at 4 K. Room temperature electronics demodulate the carriers to recover the bolometer signals, which are digitized separately and stored to disk. This readout system contributes little noise relative to the detectors themselves, is remarkably insensitive to unwanted microphonic excitations, and provides a technology pathway to multiplexing larger numbers of sensors.
In contemporary cosmic microwave background experiments, bolometric detectors are often background limited, and in this case the sensitivity of instruments can only be improved by increasing the number of background-limited detectors, and so contemporary TES receivers contain as many pixels as possible. Frequency-domain multiplexing (fMUX) is one strategy for reading out many detectors with one SQUID. For any readout system, it is important to carefully evaluate the thermal design of detector, in conjunction with the readout bandwidth, in order to ensure stable electro-thermal feedback (ETF). We demonstrate a novel technique for characterizing the thermal circuit of our detectors, using am AC-bias and the fMUX electronics. This technique is used to study the internal thermal coupling of a TES bolometer. We illustrate how the insights gathered by this technique have been instrumental in improving the stability of our multiplexed detectors for the south pole telescope (SPT).
To enable the next-generation of bolometric cameras, we are developing the microwave SQUID multiplexer (μMUX). Upcoming receivers such as Simons Observatory, CCAT-prime, BICEP array, Ali-CPT, and CMB-S4 plan to instrument focal planes with 50,000-500,000 sensors. Sensor count is achieved by tiling many 150 mm-diameter densely packed detector arrays into these focal planes. The fabrication and quality of large-format bolometer arrays has been demonstrated and is now mature. In contrast, the readout technology required for next-generation receivers needs development. The sensitivity, low cross-talk, high multiplexing density, and small component size make the μMUX well-suited for this goal. In this approach, the TES signal modulates the inductance of an rf-SQUID that loads a high-Q microwave resonator. The coupled signal therefore modulates the microwave resonance frequency, which may be read out using homodyne techniques. By coupling each resonator to the same microwave feedline, many detectors can be read out on a single coaxial cable pair. The multiplexing density is in practice limited by signal bandwidth, allowable cross-talk, and the digitization bandwidth of room-temperature readout electronics. We present the design and performance of a scalable 64-channel multiplexer chip optimized for bolometric applications. We utilize a new quarter wave resonator design that increases the physical linear density by a factor of two, therefore achieving a smaller footprint for simplified detector packaging. Measurements of this design show 100 kHz resonator bandwidth, uniform 1.8 MHz frequency spacing, and an input referred current noise of 35 pA/√Hz that is well below the level of an optimized, background-limited TES bolometer. Using 8 daisy-chained and frequency scaled chips, we create a 512-channel multiplexer and use it to readout a 512 TES-bolometer array. We present the results of this large-scale μMUX demonstration including system yield, signal cross-talk, and an analysis of noise in various TES bias configurations. The result demonstrates the multiplexing density required to read out 2,000 sensors between 4-8 GHz.
The 200-micrometer to 3-mm wavelength range has great astronomical and cosmological significance. Science goals include characterization of the cosmic microwave background, measurement of the Sunyaev-Zel'dovich effect in galaxy clusters, and observations of forming galaxies. Cryogenic bolometers are the most sensitive broadband detectors in this frequency range. Because single bolometer pixels are reaching the photon noise limit for many observations, the development of large arrays will be critical for future science progress. Voltage-biased superconducting bolometers (VSBs) have several advantages compared to other cryogenic bolometers. Their strong negative electrothermal feedback enhances their linearity, speed, and stability. The large noise margin of the SQUID readout enables multiplexed readout schemes, which are necessary for developing large arrays. In this paper, we discuss the development of a large absorber-coupled array, a frequency-domain SQUID readout multiplexer, and an antenna-coupled VSB design.