Error analysis in high-accuracy digital measurements

Knowledge of the upper limits of errors in digital sampling is very important in high-accuracy measurements, for these errors are significant elements for establishing measurement quality. In this paper, we carefully go through the theoretical background necessary for an in-depth understanding of the main sources of errors in high-accuracy measurement of harmonic components. We assess the significance of each error source both qualitatively and quantitatively, and also the efficiency of error-reduction strategies. Theoretical results are backed by simulations and also experimental results obtained with digitally sampled data. We discuss the aliasing- and integration-error interplay due to the lowpass filtering performed prior to digital sampling. We also discuss the non-causal integration model often assumed for sampling devices, e.g., digital voltmeters, and how the correction term shall be used with periodical signals in practical measurement setups.

► Main errors sources in high accuracy measurements. ► Simulation results and measured data are compared to formulas found in literature. ► Aliasing and integration errors can be very significant. ► Jitter and quantization errors were, in the cases studied, less significant.