On the one hand, this includes larger long-wave figure errors of up to several micrometers, as well as medium-frequency errors (MSFE) in the frequency range of 0.1 to 1 per millimeter. The MSFE error images contain structures in the form of spokes and as so-called "onion rings". In our example for a 22 mm Forbes asphere, both error patterns are contained slightly. The surface topography shows a low-pass filtered image that corresponds to the footprint of a scanning beam of about 1 mm. Just as we get with our ELWIMAT-V-SPOT 46/40 F#5.6 with a corresponding attachment lens f' = 200 mm on a nominal test object radius of around 20 mm. The amplitude of the IRR (irregularity function) is approx. 75 nm after correction of the SAG portion. The indicated spokes show an increased omega-12 portion. A kind of double sobrero is layered radially. The question is, how this non-ideal asphere surface can be subjected to a "proper" centering measurement and "clean" centering evaluation. As the attentive observer can easily see, the center of the lens has an asymmetrical dip with a diameter of around one to two millimeters. In addition, this dip is slightly offset to the center (estimated 50-100 µm?). How is centering measured here? Which parts of the asphere are considered? What role does the apex play when point sensors cannot resolve it with the required resolution of 1 wsec (5 nm per 1 mm rise length) and autocollimators have a spot diameter (footprint) of at least 3 to 5 mm at the apex and can't resolve it locally?
Join the discussion. We look forward to hearing about your approaches and proposed solutions.
