Test Engineer Texas Instruments, TX, United States
Parametric (analog) IC specifications are most often written with test conditions specified in English and results specified with a minimum and maximum spec or typical result. Various disciplines must then interpret (and mis-interpret) the specification and convert this into procedures, test code, and limits. Verification engineers, bench validation engineers, and ATE engineers must all repeat this similar effort to map the system definition to their unique domains, often times leading to differing fundamental test conditions and mis-correlation of the same parameters. By defining an abstract test and parametric specification definition, which is fully platform agnostic, and creating ATE and validation software as an implementation of this interface in the sense of an object-oriented software development, parametric test synthesis becomes not only possible but preferable. This allows ATE engineering efforts to focus on efficiency and fine-tuning of measurements instead of interpretation of specification. Such a system would empower ATE vendors to make sweeping and fundamental changes to their platforms and software infrastructure without requiring undue customer effort to convert solutions, and enable seamless comparison of multi-disciplinary data and supportive co-simulation. The need for, benefits of, and potential requirements of such a standard test definition language are explored in this paper.