Test Vector Based Multiple Soft Faults Detection in Linear Analog Circuits with Hardware Implementation

A method to detect multiple soft faults in linear analog circuits using test vectors is proposed in this paper. The circuit under test (CUT) is simulated used Modified Nodal Analysis (MNA) and with the knowledge of circuit topology and the component values, the test vectors associated with each components of the CUT and diagnosis variables are derived. In real time fault detection, the dependency of test vector to component values and tolerance limits the fault detection process in analog circuits. To solve this issue, test vectors are generated for nominal values, upper bound and lower bound values of the components of CUT and the fault detection is performed. Hardware implementation and verification of benchmark circuits such as Sallen Key Band Pass Filter and Linear Voltage Divider shows the robustness of the proposed approach in real time testing.