Performance advantages of buffered mode operation of HxRG near infrared detectors

Autor: Bezawada, Naidu, Ives, Derek, Alvarez, Domingo, Serra, Benoît, George, Elizabeth, Mandla, Christopher, Mehrgan, Leander
Rok vydání: 2020
Předmět:
Zdroj: Proc. SPIE 11454, X-Ray, Optical, and Infrared Detectors for Astronomy IX, 114543J (13 December 2020)
Druh dokumentu: Working Paper
DOI: 10.1117/12.2562635
Popis: The Teledyne HxRG detectors have versatile and programmable output options to allow operation of them in a variety of configurations such as slow unbuffered, slow buffered, fast buffered or unbuffered modes to optimise the detector performance for a given application. Normally at ESO, for low noise operation, the detectors are operated in slow unbuffered mode. Whilst the slow unbuffered mode offers a simple interface to the external preamplifier electronics, the detector operation in this mode can suffer from reduced pixel frequency response and higher electrical crosstalk between the readout channels. In the context of the detector systems required for the first generation instruments of the ELT (MICADO, HARMONI and METIS), an exercise was undertaken to evaluate the noise, speed and crosstalk performance of the detectors in the slow buffered mode. A test preamplifier has been designed with options to operate a H2RG detector in buffered or unbuffered and with or without using the reference output, so a direct performance comparison can be made between different modes. This paper presents the performance advantages such as increased pixel frequency response, elimination of electrical crosstalk between the readout channels and the noise performance in the buffered mode operation. These improvements allow us to achieve the same frame readout time using half the detector cryogenic electronics and detector controller electronics for the ELT instruments, which significantly reduces the associated cryomechanical complexities in the instrument.
Comment: Proceedings of the SPIE Astronomical Telescopes and Instrumentation conference 2020
Databáze: arXiv