A/D Converter INL and DNL Estimation with a Distorted Sinusoidal Stimulus Signal

Abstract

It is understood that the high precision and accurate test signal is necessary in the traditional analog to digital converter (ADC) testing. In this paper, a new method for estimating the differential nonlinearity (DNL) and integral nonlinearity (INL) of ADCs with a distorted sinusoidal stimulus is proposed. By means of a simple voltage divider circuit, let the original and divided distorted stimulus signals respectively fed into a measured ADC. Considering that the distorted stimuli are changed by the nonlinearity of the measured ADC, a Fourier series with unknown parameters is used to describe the measured ADC transition points. Since the original and divided distorted stimulus signals are used to measure the same ADC transition points, the relationships between the Fourier series with unknown parameters when the two different stimulus signals are respectively inputted to the ADC are set up. Based on the relationships, the methods for estimating the unknown parameters of the Fourier series and then the INL and DNL of the measured ADC are proposed. Both simulation and experiment results verify the effectiveness and robustness of the proposed approach.