On optimization of multi-cycle tests for test quality and application time

Abstract

Multi-cycle scan-based tests allow more faults to be detected by keeping the circuit in functional mode for more than one clock cycle. Optimizing a multi-cycle test set can improve test quality and/or test application time. It is also possible to capture the primary outputs of a circuit multiple times between the scan operations. This ensures that if a fault is detected at the primary outputs, increasing functional clock cycles of the test does not cause loss of detection of that fault. This paper presents a procedure that produces a multi-cycle test set by optimizing a single-cycle test set for fault coverage and test application time while considering stuck-at, bridging and transition faults at the same time.