Software Defined Radio Based Readout of Microwave SQUID Multiplexed MMC Arrays

14 Sept 2017, 08:30
25m
Thimann I lecture hall (UCSC)

Thimann I lecture hall

UCSC

Thimann I lecture hall
Oral Systems, Planning, Installation, Commissioning and Running Experience Systems, Planning, Installation, Commissioning and Running Experience

Speaker

Oliver Sander (KIT - Karlsruhe Institute of Technology (DE))

Description

Metallic magnetic calorimeters (MMC) are new cryogenic detectors that offer a high resolution of single eV, a signal rise time of below 100 ns, a dynamic spectrum of several 10 keV and an almost optimal linearity. MMCs are of high interest for many experiments, such as dark matter detection or neutrino mass specification. Since pixel arrays of the sensor are read out at GHz-Frequency and each single pixel offers a fast rise time, a complex analog and digital readout system is required. This contribution will give an introduction to MMC technology in combination with the implemented readout electronics.

Summary

Detection of single particles with energies of a few eV up to several MeV plays a prominent role in many areas of physics. Conventional detectors such as semiconductor detectors or crystal spectrometers either offer a high dynamic range with limited resolution or vice-versa. In contrast, cryogenic detectors such as metallic magnetic calorimeters (MMCs) combine high spectral and temporal resolution while covering a broad energy range at the same time. Modern magnetic calorimeters as used for soft X-ray spectroscopy offer an energy resolution of 1.6 eV at 6 keV particle energy, a signal rise time below 100 ns, an energy bandwidth of several 10 keV and an almost ideal linear detector response. For the readout of MMCs, superconducting quantum interference devices (SQUIDs) are used since they provide a high system bandwidth and low noise. The latter two properties are mandatory for detectors with a high temporal and spectral resolution.

In addition to these, many experiments require spatial resolution to determine the location of an event or to collect a lot of events for high statistics. Both requirements can be fulfilled by using multi-channel detector systems, which contain a lot of independent pixels. But at the same time, the readout of such multi-channel detector systems turns out to be challenging. Microwave SQUID multiplexing as previously introduced by Irwin et al. for reading out arrays of superconducting transition edge sensors turns out to be a very promising approach. Here, non-hysteretic rf-SQUIDs are used to modulate the pixels’ information on different carrier frequencies in the GHz range. The first part of our contribution will briefly introduce the applied MMC technology along with the highly innovative multiplexing technology.

In addition, the requirement of processing of the readout electronics imposed by MMC arrays is more demanding as compared to reading out transition edge sensor arrays due to the fast signal rise time of MMCs. For this reason, a customized readout system for microwave SQUID multiplexed MMC arrays will be presented as major in this contribution. The readout system is based on software defined radio. All digital processing is implemented on an FPGA. Fast Digital-to-Analog-Converters (DAC) and Analog-to-Digital-Converters (ADC) are used for creating and digitizing the MHz frequency comb which is sent to the multiplexer and modulated according to the actual state of the detectors. A customized high-frequency front-end electronics performs the up- and down-mixing of the frequency comb to the targeted frequency range of 4-8 GHz. Each of the three parts has been developed, implemented, integrated and evaluated. The first version of the system was successfully used to interface a 64 pixel MMC detector array. This solution allowed for the very first true multiplexed readout of a metallic magnetic calorimeters, i.e. to record events on various independent channels in parallel, by just using one pair of coaxial cables into the cryostat. Moreover, we will show, how this system architecture can be extended to the readout of thousands of channels in parallel. Eventually, we target the simultaneous readout of up to 100 k channels with this approach.

Primary authors

Oliver Sander (KIT - Karlsruhe Institute of Technology (DE)) Mr Nick Karcher (Karlsruhe Institute of Technology) Dr Sebastian Kempf (KIP - University Heidelberg) Mr Mathias Wegner (KIP - University Heidelberg) Dr Oliver Krömer (Karlsruhe Institute of Technology) Matthias Norbert Balzer (KIT - Karlsruhe Institute of Technology (DE)) Prof. Christian Enss (KIP - University Heidelberg) Marc Weber (KIT - Karlsruhe Institute of Technology (DE))

Presentation materials