A Digitally Programmable A/D Converter for Smart Sensors Applications

Nonlinear analog to digital conversion in smart sensor applications is an important topic since signal digitalization and linearization can be performed in a single step nearby the transducer. In this paper, a double pulse width modulated (PWM) scheme for nonlinear analog-to-digital conversion is presented. Calibration or auto-calibration data stored in the smart sensor´s memory define the nonlinear profile characteristic of the transducer and provide the required data to obtain the inverse function of the analog-to-digital converter (ADC) transfer curve. Basically, as a function of the transducer´s nonlinearity degree, the input voltage range of the ADC is segmented in a continuous set of sub intervals and for each of these subintervals a second order correction term, based on a PWM A/D conversion, is used to obtain a linear characteristic for the smart sensor. Additional advantages of this method results from its easy implementation in low-cost microcontrollers that include generally comparator inputs and PWM outputs. A flexible and programmable A/D conversion solution can be dynamically adapted to variations of the transducer´s nonlinearity profile and an increased resolution can be achieved at expenses of a lower conversion rate. Some MATLAB simulations and experimental results obtained with a square-root airflow transducer are presented in the final part of the paper