When it comes to measuring large uncoated glass components, the Ritchey-Common test has yet to find acceptance, although it is a well-established interferometric technique for determining the surface shape of large (up to 1 m) reflective flats. This may be about to change.Engineers at Veeco Instruments Inc. in Tucson, Ariz., have reported successful characterization of uncoated optical surfaces with the test using a commercial interferometer and custom analysis software. This method can provide a cost-effective strategy for manufacturers to control the quality of their optics before and after the coating process.Researchers have reported that the Ritchey-Common test, used with a commercial interferometer and custom analysis software, enables measurement of large uncoated optics. The Ritchey-Common procedure allows the use of a small-aperture, phase-shifting interferometer for measuring much larger flats, instead of a custom device with a larger aperture than the object under test. There are four steps: measuring the optical distortion of the system, determining the incidence angle of a test flat, measuring the empty cavity as a reference and using software analysis to determine the shape of the optical component.The process directs a test beam through a transmission sphere, such that its diameter increases with distance. The flat to be tested is inserted in the optical path at an angle, and a return sphere slightly larger than the flat reflects the beam through it. This combines with a reference beam to produce interference fringes. A series of measurements are taken, from which the analysis software extracts the optic's surface shape."The key to this method is to achieve good contrast between the test and reference beams," said Sen Han, senior optical engineer for Veeco's optical and industrial metrology business unit. For measuring a reflective flat, the return sphere typically remains uncoated. For measuring a bare flat, however, a reflective coating must be applied to the return sphere, to return sufficient light to the interferometer."With a 95 percent coated return sphere, the reflectivity of the test beam is only 0.15 percent," he added. "We therefore apply an antireflective coating to the transmission sphere to match the intensity of the reference and test beams. Using this arrangement, we are able to measure uncoated flats with good results."