Mirror testing

an alternative to the Foucault test

Like most people my introduction to this subject involved the Foucault test method. It still amazes me that features measured in fractions of a wavelength of light can be detected on a glass surface with a device that can be cobbled together from a torch and a couple of razor blades. However, for figuring purposes the method does have certain shortcomings. It can be difficult to judge the positions of subtle shadows and hence determine the correct knife-edge settings, and the method can be prone to bias in taking those readings. My approach to testing now involves the use of the Foucault for detecting general errors but a different method for final figuring of the mirror. This employs the Ronchi grating in two steps, a 'bench' test and a 'field' test. The latter can be used as a fairly simple star test for a telescope, do-able by a beginner.

1. The 'bench' test.

    This method is fully described by Mel Bartels, so I will not repeat it here in detail. It is quite simple and involves the use of a program written by Bartels called Ronchi.exe. The program generates a set of six images (NOT those shown below) showing how the Ronchi lines should appear for a given mirror at certain positions of the grating. The user then simply makes a direct visual comparison between the generated and observed patterns. And that's it! Even on its own this has the potential for producing quite decent mirrors, but for greater performance should be combined with some form of star (i.e field) test.

2. The 'field' test.

   This is described by Gerald North (in Advanced Amateur Astronomy), and is also a simple test to perform. It requires the use of a Ronchi grating eyepiece.

    A bright star is located in the telescope and the optical eyepiece replaced with the Ronchi grating eyepiece. The focuser is adjusted so that four lines are visible across the circle of light that is seen. The shapes of the lines can then be compared with the patterns given here.

A: Well corrected mirror.

B: Overcorrected (i.e. too deep).

C: Undercorrected (i.e. not deep enough).

D: Turned down edge, indicated by the hooked ends of the lines.

E, F: Raised and depressed zones.

N.B. It is important that the mirror is at the same temperature as the surroundings otherwise patterns like B or C may occur but only be temporary.

    Obviously A is what you want to see. But do not be too alarmed if this is not the case. Except in the case of turned down edge. If you have this then correct it or mask it out. The latter is simple enough and just involves fixing a ring of thick matt black-painted card about 5mm above the edge of the mirror, which has an aperture of 6-12mm less than the mirror.

    B,C,E and F may not seriously impair image quality, if they are not too severe. If in doubt try this simple test:    

N.B. These comments assume accurate collimation, equilibrated mirror temperature, clean optics etc.

Target a planet (preferably under good seeing conditions, i.e. when the image frequently appears steady) and examine the image under low (10-15x/in of aperture), medium (20-25x/in) and high power (30-40x/in).

    If the image is poor under low power then the telescope is worthless.

    Image good at low power but poor at medium power - poor optics.

    Good  at medium though poor at high power - optics OK, a satisfactory telescope.

    Good at high power - great telescope!