A Novel Design Evaluation Concept Applied to Switched-Current Algorithmic A/D Converters

Currently used design evaluation approaches of A/D converters are often based on ad-hoc methods using the designer´s intuition and previous experience. The verification process carried out in this "trial and error" way usually takes an exhaustive time, with no possibility of subsequent design optimization and device calibration. In this paper we propose a novel design evaluation concept suitable for Nyquist-rate A/D converters capable to overcome all the above-mentioned problems. The core of our methodology is a root-cause identification of the error sources present in the ADC structure, evaluating the performance degradation by static non-linearity in a series of consecutive steps. For this purpose, the recently introduced technique of LEMMA (linear error mechanism modeling algorithm) is used, oriented to systematical decomposition of the ADC structure. The proposed methodology was successfully verified on a design example of switched-current (SI) algorithmic ADC, as it is introduced and thoroughly described in this paper