Bridging, transition, and stuck-open faults in self-testing CMOS checkers

Abstract

The consequences of bridging, transition, and stuck-open faults in self-testing checkers designed only for single stuck-at faults are examined. A methodology for design that guarantees that the checkers will be self-testing in the presence of bridging, transition and stuck-open faults is established. This methodology is applied to several implementations of self-testing checkers. Simulations confirm that these checkers are self-testing in the presence of bridging, transition, and stuck-open faults. The problems associated with testing the checkers in the presence of non-stuck-at faults and the problems that result from reducing the number of checker outputs from two to one are discussed. It is shown that self-testing checkers designed for stuck-at faults will remain self-testing in the presence of nonclassical faults.<>