Time-Pulse Converter for Capacitive Parametric Sensors

Abstract

The article is devoted to the urgent issue of metrological justification of circuit solutions choice for the construction of time-pulse converters of capacitive parametric sensors used in modern industrial automation systems and control measurement technology. The paper provides a detailed comparative analysis of three common hardware implementation options for measuring channels: based on discrete voltage comparators, a specialized integrated single-vibrator SN74ALS123 and a precision integrated timer of the NE555 series. The feasibility of using such solutions as a reliable and much cheaper alternative to specialized integrated capacitor-to-digital converters (CDC), which are often too complex for typical engineering tasks, is substantiated.

The research is based on the development of a unified mathematical model of the measuring conversion function, which is based on the analysis of the exponential transient process of the charge of the time-determining RC circuit. The mathematical similarity of the transformation functions for the entire studied group of circuits has been proven, which makes it possible to use single signal processing algorithms. Using computer modeling in the Maple analytical environment, a metrological analysis of the influence of technological tolerances of passive components and variations in internal parameters of microcircuits on the final measurement result has been performed. Three-dimensional characteristic surfaces have been constructed that illustrate the dependence of the output digital code on the sensor capacitance and hardware parameters of the circuit.

It has been established that when using a quantization clock frequency of 100 MHz, the relative sampling error does not exceed 0.09% in the range of about 1 nF, which is unreasonably small compared to the errors of the analog part. It has been determined that the dominant factor of the total error is the technological spread of the parameters of the reference resistor (±5%) and the microcircuit's own proportionality coefficient (±2%). To eliminate these effects, a two-level software correction method is proposed, which is based on the procedure of individual calibration of each device instance. The proposed approach provides high measurement accuracy when using the available element base.