There should not be any issues regarding the use of the steel disc in place of the ruby ball. The calibration of the disc will take care of all but the static / magnetism issues. If you have issues with this maybe you should stay with the ruby but go with a larger diameter. You should not have an issue with cosine errors if you are measuring the features as a circle that automatically places vector touches on the artifact being measured. Any errors encountered with an out of round part given the size of the parts you are measuring will be undetectable within the accuracy range of any CMM