Soft Fault Diagnosis of Analog Circuit Using Transfer Function Coefficients

Abstract

A method to identify parametric faults occurring in analog circuits is proposed in this paper. This method uses transfer function coefficients to identify faults in analog circuits as these coefficients are sensitive to the parameters of the circuit. Using Monte Carlo simulation each parameter of the circuit is varied within its tolerance limit and the minimum and maximum values of each coefficient are found for faulty and fault free conditions. While testing the coefficients are found for the injected fault and if all coefficients are within the predetermined bound limits of any fault conditions then the circuit is confirmed to have that particular fault. In the same manner all single faults occurring in the analog circuits are identified. The proposed method is illustrated through second order sallenkey band pass filter circuit. Results have confirmed that the proposed method can diagnose single parametric faults in analog circuits efficiently.