A Foucault tester or knife-edge tester is used to measure the radius of curvature of a mirror or sections of a mirror. To focus the light from a star, the primary mirror in a telescope should have the shape of a paraboloid. Foucault realized that a paraboloid mirror could be approximated as a series of concentric rings with slightly different radius of curvatures. He devised a knife-edge test to measure the radius of curvature of a mirror, or of sections of a mirror. A more complete discussion of knife-edge testing can be found in the Waineo Null Test page and the references therein.

My first Foucault tester followed the design outlined by Berry (1) and worked well. I decided to build another that could also be used for a Caustic test. (2) I find a Foucault test more convenient than a caustic test and use it much more often. The Foucault/Caustic tester was built following the directions in articles by Schroader(2) and by Macintosh. (3) If you read Macintosh's directions, he harps on the "necessity" of aligning the two axes perpendicular. I believe he is wrong, and that modest misalignment ( a degree or three) of the two axes has negligible impact on the accuracy of the tester. Macintosh has some good advice, but the Schroader design is more down to earth. One of the reasons I usually use the setup in Foucault mode is that the requirements for precision in the transverse direction for a caustic test are ca. 0.0001". Most often I do my testing in a room with a wooden floor, and a cat walking next to the testing table will lead to transverse displacements this large. If you are serious about caustic testing, you should be working on a very steady table on a cement floor.

The tester is has a fixed light source. It is a sanded down LED behind an adjustable slit. The slit was prepared as described Texereau.(4) The tester could be reconfigured to use a moving slit, but I have not done it yet.

If you decide to use a caustic test, you can work up the data with Sixtests. I usually use the program tex to work up my Foucault data, because it is available for Macintosh computers. Be aware that the Schroader analysis of the caustic test is only correct if you have a moving pinhole source. When the pinhole is fixed, there is a correction in the formulas that should be considered.(5) For amateur sized optics the correction is rarely relevant.

Here are some pictures of my Foucault/Caustic tester.

References

(1) R. Berry, Build Your Own Telescope, Second Edition, 1994, Willmann-Bell, Chapter 11.

(2) I. H. Schroader, "The Caustic Test," in Amateur Telescope Making, Volume 1, A. G. Ingalls, Editor, Willmann-Bell, 1996, pp. 251-284.

(3) A. Mackintosh, "Caustic Testers," in Advanced Telescope Making Techniques, Volume 1: Optics, A. Mackintosh Editor, Willmann-Bell, 1986, pp. 33-38.

(4) J. Texereau, How to Make a Telescope, Second Edition, 1984, Willmann-Bell, pp. 59-85.

(5) D. P. K. Banjerjee, R. V. Willstrop, and B. G. Anadarao, "Improving the Accuracy of the Caustic Test," Applied Optics, 1998, 37, 1227-1230.