A Fabrication Process Variation Based Approach to Evaluate Design-for-test Techniques

One of the hurdles that prevent the wide acceptance of analog/mixed-signal design-for-test techniques is lack of a realistic and practical evaluation methodology. In this paper, a method to assess the quality of a design-for-test design (together with the associated test set) is proposed. Based on fabrication process variation information, our design-for-test evaluation criterion considers the inevitable adverse effects of fabrication process imperfection on both the functional and design-for-test circuits, and therefore will correlate better with the manufacturing test quality than past proposals. To validate our method, a flash ADC with a built-in ramp generator is used as an ex-ample. The proposed approach shows that the original designfor-test design fails capturing dynamic defects, and a redesign of the ramp generator or the test set is necessary to enhance the manufacturing test quality.