Design of partial nulls for testing of fast aspheric surfaces

Standard interferometric testing of aspheric surfaces makes use of a null optic that is matched exactly to the surface under test. This leads to part specific null lenses having to be designed, manufactured and tested for every aspheric surface to be tested. This paper discusses design issues associated with testing a range of fast aspheric optical surfaces with a single partial- or pseudo-null lens. Collecting the light that is reflected from a range of fast aspheric surfaces becomes one of the major design concerns. The partial null should also provide a reduction in the wavefront slope. The slope of the wavefront, or more specifically the fringe frequency created by the interference of the test and reference wavefronts, must not exceed the maximum measurable fringe frequency of the detector used in the system. Furthermore, since the null condition has been removed from the interferometer, light no longer travels the same path to and from the test surface. This situation leads to retrace errors introduced into the test wavefront by the interferometer. These errors must be calibrated as part of the measurement process in order to recover the aspheric test surface. The ability of the system to be calibrated must be considered in the design of the partial null.