Share Email Print

Proceedings Paper

Partially coherent illumination in interferometry for optical testing
Author(s): Johannes Schwider
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

The production of high quality optical components heavily relies on interferometric testing methods /1/. Before the invention of the laser proof glass methods dominated in the production line of the optical shop. Only in scientific labs or special testing laboratories Fizeau- and Twyman-Green interferometers were used. The reason was that the illuminating sources were spatially incoherent spectrum lamps or in the best case isotope lamps /2/. In order to obtain contrast-rich fringe patterns one had to stop down the light sources or in more involved cases to adapt the reference arm-length /3/. The big advantage of the fringes with partially coherent illumination was their smoothness enabling very high measuring accuracy. But extended incoherent and monochromatic sources restricted the interferometer design to the most simple devices as the Fizeau- and the Michelson-interferometer and with special precautions also to the Twyman-Green interferometer. With the invention ofthe laser interferometry became a fool proofmatter. The high spatial and temporal coherence of single mode laser radiation together with the high luminosity ofthe laser allows for total design freedom. This is of course a great advantage on the one hand but on the other the high degree of coherence is responsible for the disturbances caused by small scattering particles in the ray path which cause so called dust diffraction patterns. Through this effect the measured phase is disturbed by stochastic phase variations which limit the wavefront smoothness to the /1OO-level. Although a well designed interferometer will be almost free from the latter problem in some regions ofthe field of view in others the disturbance will remain. In two-dimensional measurements of phase distributions as in optical testing applications these disturbances can not be averaged out as in the case of simple length measuring devices /4/. There will always be a repeatability problem with varying adjustments of the interferometer. In modern interferometers CCD-detection is state of the art and therefore measurements with these devices show the problem very clearly. As will be shown, the repeatability problems with two-dimensional shape measurements result from the dimensionality and not from the ultimate sensitivity. With zero-dimensional length measurements the accuracy can be of the order I O or better since one averages over many dust diffraction patterns with the detecting area. In contrast to this two-dimensional measurements only reach iO to i04 accuracy levels with some effort.

Paper Details

Date Published: 17 August 1999
PDF: 12 pages
Proc. SPIE 3745, Interferometry '99: Applications, (17 August 1999); doi: 10.1117/12.357761
Show Author Affiliations
Johannes Schwider, Univ. Erlangen-Nuernberg (Germany)

Published in SPIE Proceedings Vol. 3745:
Interferometry '99: Applications
Werner P. O. Jueptner; Krzysztof Patorski, Editor(s)

© SPIE. Terms of Use
Back to Top