The future of frequency multiplexed electronics for RF

The future of frequency multiplexed electronics for RF superconducting detectors Gustavo Cancelo Detector R&D meeting April 19, 2019

f. MESSI: Fermilab’s frequency multiplexed electronics for RF superconducting detectors (2016 -2019) To MKID Up conversion, amplification, attenuation and filtering from MKID Down conversion, amplification, attenuation and filtering RF out IF in RF in IF out 10, 000 detectors LO ROACH 2 Fermilab electronics To MKID from MKID 2 To/from ROACH 2 High channel density: 1, 000 to 4, 000 channels/board. Low power: 45 Watts/board. Low cost: $7000/board + Roach 2 Gustavo Cancelo | Scalable 10 to 20 Kilo-pixel MKID Signal Generation and DAQ for Cosmology 11/30/2020

The future of frequency multiplexed electronics for RF superconducting detectors: 20 K MKID system for Subaru An entire rack with 20 f. MESSI + 20 Roach 2 boards can be replaced by 1 board with 2 Xilinx RFSo. C FPGAs. The evaluation board (bottom-left) allows us to do the proper R&D that will make us take advantage of the next generation of experiments using superconducting detectors (SCD). In particular SCD that require: • High number of multiplexed channels, up to 4 k ch. or more (e. g. CMB-S 4 and DE-MKIDs). • SCD RF detectors (e. g. CMB, Axions, MKIDs, QIS). • Ultra low latency: Control of SC qubits for QIS and Axion detection. • Low #ch but high BW (e. g. SCQ) Highlights of RFSo. C FPGA: 16 x 12 bit ADCs at 4 Gs/s each. 16 x 16 bit DACs at 6. 5 Gs/s each. 4 core and 2 core ARM processors (Linux). 10 x the logic and memory of a Virtex 7 FPGA. Built in interfaces 100 Gb ethernet, USB, I 2 C, etc. Fermilab has started a by weekly collaboration meeting with US labs and Univ. in the country 3 Presenter | Presentation Title 11/30/2020