2nd Advanced Mirror Making Seminar

Foucault tester made by Richard Taylor

19 inch Mirror brought by Richard Taylor

Ronchigram of Richard Taylor's 19 inch mirror showing astigmatism

Foucault tester made by Tom Green

John Wall and Peter Wise

Telescope by Peter Wise. (All spherical optics)

24^th January 2004

About 29 interested parties gathered together for the 2^nd seminar on mirror making, this time the theme for the day was testing. As with last years meeting the agenda was loose and discussions often drifted of topic, everyone seems to prefer it like that!  Below is a summary of some of the day’s events, not necessarily in chronological order.

The meeting was opened by Steve Taylor of the Astronomy Centre, who was again hosting the meeting and taking on much of the organisation, well-done Steve.

The opening session included a discussion on ‘how good is good enough’, which proved to be a very difficult question to answer. I think a consensus in the end was  'good enough to take advantage of those one or two nights per year when seeing was exceptional'. The optics need to be better than the best seeing you get…but not hugely better. Peter Wise commented that occasions when seeing was better than .5 arcsec were rare.

Steve went on to demonstrate a piece of software called Aberrator. This software simulates the effects of errors on mirrors.

http://aberrator.astronomy.net/

John Wall gave an interesting account of using an oil flat to test a large mirror by autocollimation. He had tried mercury but it was much more susceptible to vibration than oil.

Es Reid presented a summary of the James/Wanieo test. Apparently, Bill James had proposed this test quite a few years before Wanieo popularised it. The test involves testing a paraboloid with a spherical mirror.  Es went on to talk about the Offner null test, which in one form consists of two lenses. The test is similar to the Ross and Dall tests but is superior because it takes into account higher order spherical aberrations unlike the Ross and Null tests. A discussion followed concerning the Hubble mirror which was tested using an Offner test. Es pointed out that the mirror was tested with a lens based system initially and finally with a mirror based version of the Offner test. It was this latter setup that had the flaw; the spacing of the optical elements was in error. It was pointed out that if the telescope had been tested visually the error would have been obvious. Es proposed that the lens version of the Offner test should be used to test the Astronomy Centre 40 inch mirror.

Peter Wise showed some computer generated spot diagrams for his all-spherical design. Spot diagrams were shown for various incarnations of the system. Peter is making an 8 inch version of the system commercially and had brought a finished prototype telescope with him.

 Dave Thompson and the Liverpool Sidewalk Astronomers discussed the testing of the 30 inch mirror (1.5  inches thick), made by them. A couder mask was utilised in conjunction with the Foucault test. By all accounts the mirror has turned out well and is producing good images.

Richard Taylor had travelled up to the meeting (Heywood, Lancs.) from Bristol bringing with him a 19 inch mirror he had been working on. The glass was plate, about 1.5 inches thick. The mirror was set up under test and was scrutinised by many at the meeting. The mirror showed some astigmatism, not unexpected with a large thin mirror. Norman Oldham (Oldham Optics) commented that if a blank is not circular astigmatism would be seen, others commented on the need for minimal wedge (difference in edge thickness) in a mirror blank. Richard's mirror also had many fine scratches/sleeks. John Owen commented  that they were probably due to hard pitch.

 Steve Taylor demonstrated the software FringeXP that is available as a free download. The software is for reducing the results of interference testing using an interferometer. A discussion on the various types of interferometer followed. Mention was made of Peter Ceravolos website (www.ceravolo.com) which outlines the building of an interferometer.

 I had the pleasure of presenting a short piece on the Ross null test. This is a test not dissimilar to the Dall null test. A plano convex lens is used to introduce spherical aberration equal to, but opposite in sign to the mirror under test. The Ross test uses a large lens, light passes through the lens twice, once on it’s way to the mirror under test, and once on its return. This avoids astigmatic effects that are present in the Dall test in which the light source and knife edge (or Ronchi ruling) are separated by at least the diameter of the test lens and it’s housing.

Throughout the day many useful contributions were made by those present, could I in particular mention: Peter Drew, Steve Taylor, Norman Oldham, Peter Wise, John Owen, John Wall, Es Reid, Dave Thompson and others, apologies to those whose names I did not know.

During the breaks I was able to chat to some of the delegates, a couple of interesting points arose concerning flats. When used as secondary mirrors in the Newtonian configuration the feeling was that surfaces do not have to be excellent. Some degree of curvature is acceptable, but the surface has to be smooth. When used at high power a smaller area of the flat is used, it is this area that needs to be of higher quality. Concerning the use of large flats for autocollimation testing, again some degree of curvature is acceptable, but again the surface needs to be smooth.

Thanks go to Lizzy Taylor for her excellent catering arrangements and to all those involved in organising what turned out to be another excellent day for ATM’s. I look forward with eager anticipation to the next meeting!