ADC non-linear errors correction in thermal diagnostics for the LISA mission

TL;DR

This paper addresses the correction of ADC non-linear errors in thermal diagnostics for the LISA mission, demonstrating methods to reduce excess noise caused by ADC transfer curve non-idealities in low-frequency temperature measurements.

Contribution

The paper introduces two methods using out-of-band dither signals to mitigate ADC non-linearity errors in thermal measurements for LISA.

Findings

Successfully reduced ADC non-linearity noise in temperature measurements.

Achieved noise reduction for temperature ramps up to 10 uK/s.

Improved measurement accuracy in low-frequency thermal diagnostics.

Abstract

Low-noise temperature measurements at frequencies in the milli-Hertz range are needed in the LISA and LISA PathFinder (LPF). The required temperature stability for LISA is around 10 uK/sqrt(Hz) at frequencies down to 0.1 mHz. In this paper we focus on the identification and reduction of a source of excess noise detected when measuring time-varying temperature signals. This is shown to be due to non-idealities in the ADC transfer curve, and degrades the measurement by about one order of magnitude in the measurement bandwidth when the measured temperature exhibits drifts of uK/s. In a suitable measuring system for the LISA mission, this noise needs to be reduced. Two different methods based on the same technique have been implemented, both consisting in the addition of dither signals out of band to mitigate the ADC non-ideality errors. Excess noise of this nature has been satisfactorily reduced by using these methods when measuring temperature ramps up to 10 uK/s .