I work in the QA department of our company. We are advancing more and more into small parts (tolerances +/-.002). We have an OGP Zipscope 250 video microscope for checking samples on our incoming inventory. Currently, we are in disagreement with the engineering dept. over our choice of measuring tools. There is discussion on whether we can use the automated vision system to verify parts as good or bad. From my perspective, I trust the vision system over everything else. The other options we have are CMM, Optical comparators and gage pins/microscopes. The parts we are testing are rubber and obviously susceptible to contact measurements. My question comes down to two areas. How do you prove the accuracy of the vision system to engineering? How, in measuring out of round rubber seals, do you determine diameters to begin with? In terms of the out of round dilemma, best fit may not be the best way to go, so I would like to hear what anyone else has to say regarding any of these issues. Has anyone else run into this kind of problem and what is the general practice in industry regarding these kind of measurements. Thanks for any assistance.