Phase measuring interferometers can provide extremely repeatable comparisons between reference and test surfaces. This thesis explores the limits to which the accuracy of measurement of flat, spherical and aspheric optical surfaces can be made to approach that repeatability. Analysis of conventional methods for separating errors in flats (the three flat test) shows that the systematic differences between repeated measurements can be predicted from the shape of the surfaces and a plausible estimate of relocation errors. Errors introduced by distortion are analysed. New methods for separation of errors associated with the reference surface and test part are introduced, together with a new method for visualising the errors. When aspheres are measured with respect to a spherical reference, non-common path errors may be substantial. A software correction method, based on the use of Zernike polynomials, is introduced for errors associated with the imaging system of the interferometer. Further errors arise from non-common path use of multi-element transmission spheres. The errors can be estimated using ray trace code, and corrections made. The uncertainties are, however, higher than required for future optical systems. Thus the concept for a next generation interferometer, not subject to the limitations identified in this work, is introduced
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:552738 |
| Date | January 1996 |
| Creators | Evans, Christopher James |
| Publisher | University of Birmingham |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.002 seconds