Glasspusher

Managed to capture this raw spectrum of Nova Per using a 10 inch Newt and a SA 100, single exposure of 90 secs. H alpha and H beta emission lines shown quite well. John

Thought I would give this thread a bump as it is 15 years since David's untimely passing. I am pleased to say that my paper celebrating David's life and achievements has been accepted for publication in the BAA Journal and will appear this year. In the mean while here are a couple of pictures, the first shows David with a 1.2m mirror for a telescope in India, the second shows the completed instrument. Thanks for looking, John

Found this chart on the web, not sure how reliable it is but you can see that your mirror is close to the red line indicating the difficulty in using a Bath interferometer for this project. Hope you find this of interest. John

As you say the advantage of using borosilicate glass would be reduced time between figuring sessions due to the better thermal properties of the borosilicate glass. Regarding supporting the convex back during working, I ground the convex back to get a smooth regular figure I then cast a support against the convex back to ensure the support matched as well as possible. Despite this and regular rotation relative to the support I still got astigmatism. I don't think a wooden support would be stable enough for a mirror of this size. The stability and uniformity of wood is a concern. The slightest deviation from a regular support can result in astigmatism. I haven't asked as yet, but what sort of surface accuracy are you hoping to achieve? John

The fire brick mould worked well and survived several firings. The 500mm mirror is plate glass; with thin mirrors like this it is not so essential to use borosilicate glass as the glass cools quite quickly being relatively thin. The mirror was finished and tested but suffered from astigmatism probably due to poor support during grinding/polishing. I should probably go back and have another go at it. The reason that I did not continued with kiln work was that the cost of running the kiln was becoming a little excessive. It was apparent that it would not represent a significant saving, if any, over a monolithic blank. The literature available on making slumped mirrors, as you are finding, is limited compared to that on monolithic mirrors which are more ‘tried and tested’ so I decided to stay with the monolithic route. John

Gluing is definitely not the way to go. Fusing and careful annealing will produce a usable result. The Hubble sandwich mirrors are of this type and many are happy with the performance. See also the work of Normand Fullum: http://www.optiquesfullum.com/optical-mirrors/ John

Yes that was Dave Thompson who was part of the group that built the 30 inch scope. The problem in using a Bath interferometer is that the resulting interferogram will have crowded fringes due to the size and speed of the mirror making analysis difficult. You will need to obtain good clean high resolution igrams with such a fast mirror. It has been a few years since I did kiln work. I made the mould from kiln fire bricks glued together with kiln cement as shown in the pictures. I sanded in the curve using a glass disk with the approximately the same curve as required. A circular self adhesive sanding disk was stuck onto the glass which was about 50% the diameter of the mould. The process is just like grinding a mirror and was continued until the correct curve depth was reached as measured with a spherometer. I have attached some images, first one is of the finished mould. Next three are the finished 20inch diameter by 25mm thick meniscus mirror. The forth one is the mirror stress test, even after a second annealing run irregular stress still present. With kiln work it is essential to do all heating and cooling at the slowest rate particularly when annealing. The final image was my attempt at fusing a 20 inch mirror using 15mm thick plate glass. In the end the cost of running the kiln was getting silly. I reverted to the tried and tested monolithic blank. Hope some of this helps. John

Just few further thoughts regarding the mirror blank. Slumping is very much in it’s infancy as far as amateur telescope makers are concerned. If you choose the slumping route you will have to consider: The cost of a kiln and how to use it effectively. How to make a suitable mould. If the mould is not accurate enough you might have to do extra grinding to get the radius required. This will result in loss of thickness which is not helpful if you are starting with what is already thin glass. Supporting the mirror during grinding/polishing, if not done properly you will end up with astigmatism. These comments are based partly on the experience of slumping and grinding/polishing a 20 inch x 25mm thick mirror. With a pre-generated monolithic blank you could cast a dental stone tool using tiles or steel nuts and begin work straight away. If you start with a flat blank and grind in the curve with a sub-diameter tool you will loose some blank thickness, this should not happen with a pre-generated blank. Such a blank can be purchased here… http://www.reginato.it/blanks.htm Alternatively, Thompson telescopes have a 760mm F3 blank listed on their website… https://www.thomsontelescopes.com/store.html Hope this helps. John

As a mirror maker with many years experience I wonder if I might share some thoughts on your ambitious project. When making a big mirror the first thing to consider is how do I test it? If you can’t test it don’t make it. Why go through all of that work in the hope that testing will be achievable? When testing mirrors the difficulty increases with larger diameter and shorter focal ratio. So a 300mm F5 mirror can be figured in a few hours or less by an experienced optician. An 800mm F3.3 is an entirely different proposition. All forms of testing require a degree of skill which is acquired over years of experience. What level of accuracy are you hoping to achieve? It is unlikely that you will end up with a high resolution planetary telescope; a low to medium power deep sky telescope is a more likely outcome. Thinking about testing you might consider: The Foucault test. The mirror could be tested at the centre of curvature with a zonal mask but this test becomes less reliable with increasing diameter and speed of the mirror. Reading the shadows is a subjective business and requires years of experience to get it mastered. http://www.jeffbaldwin.org/figure.htm The Ronchi test. Again this could be used. The test is basically qualitative but Mel Bartels amongst others, has developed a semi-quantitative version. The observed Ronchi patterns are compared to computer generated images of an ideal Ronchi pattern. The mirror is worked until the observed Ronchi pattern matches the computer generated pattern. This is done for a number of positions both inside and outside of the centre of curvature of the mirror. https://www.atm-workshop.com/ronchi-test.html https://www.bbastrodesigns.com/ronchi.html Star testing. The best method of testing a mirror is on the stars. You need a tube assembly to pop the mirror in between figuring spells. Mirror could be figured using the matching Ronchi test to get is as near as possible, final figuring being done using the star test. https://www.bbastrodesigns.com/JoyOfMirrorMaking/StarTesting.html Other ‘less common’ tests include: Double Pass Autocollimation test (DPAC). An excellent null test used extensively by professional optical shops the world over. You need a high quality optical flat, usually the same diameter as the test optics. A smaller flat might be used with care. Optical flats are very expensive items. Waineo test. This uses a spherical mirror to perform the null; the mirror can be smaller than the test optic. Again additional optics are needed on this occasion a spherical mirror of good quality. The test goes by several alternative names. https://www.bbastrodesigns.com/waineo.html Interferometry. The Bath interferometer has become popular with amateur telescope makers in recent years. Unfortunately it is not suited to large fast mirrors. http://rohr.aiax.de/Using a Bath - EN.pdf Optical testing: https://www.telescope-optics.net/testing_optical_quality.htm The Secondary mirror. A large flat secondary mirror is a very expensive item and not one which is easily made. Most telescope makers prefer to buy a secondary mirror which will be a very expensive item. I have experimented with both fused and slumped mirrors over the years. You would need access to a large (and expensive) kiln for a project such as this. There is a steep learning curve which can be very expensive due to the cost of electricity and breakage of glass. Annealing takes much longer than anticipated which means running the kiln for a numbers of days. Expensive! I think a monolithic blank would be better, as to the thickness, the thicker the better unless you have had experience in controlling astigmatism in thin mirrors. The problem with astigmatism is that in its more subtle forms it is easily missed until you come to examine a star image with the telescope. Sorry that much of this sounds a little negative but undertaking a project like this is no trivial task, many have tried, a few have succeeded. Good Luck. John

To clean the nut tool between grits I scrub with a stiff brush under running water if not satisfied with that I use a wire brush which usually does the job. With tiles or glass chunks water can ingress into the dental stone down the edges which can cause the pieces to become loose and draw in abrasives. You can see this more obviously at the edges of the tool where the tiles are exposed. Always try to set the tiles in from the edge so that they are surrounded by dental plaster and use the thickest tiles you can find, 4 mm is a little thin in my experience. John

Over the years I have tried every type of grinding tool and a nut tool is difficult to beat. I use M16 steel nuts set in dental plaster. Cover the concave glass mirror with something like greaseproof paper, place a damn round the edge, place the nuts on the greaseproof paper and carefully mix up the plaster and pour into the mould. Allow to set and that is it. During use if the tool has dried out immerse in water for a few minutes before use. Worth thinking about! Good luck with your mirror! John

Hello, I have slumped my own blanks in a kiln but gave up due to the expense of running the kiln. To properly anneal a large blank takes many hours of controlled cooling which is expensive. You then have the problem of supporting the meniscus blank during grinding which is not straight forward. It is more economical to go the monolithic blank route which is well tried and tested. Hope this helps. John

I could not agree more, well done Robin! John

Saganite, Thanks for your input, you can find more about the UKIRT here: https://sites.google.com/site/grubbparsons/home/grubb-parsons/telescopes-made-by-grubb-parsons/150-in-mirror-ukirt-hawaii-1976 UKIRT was a very successful telescope, unfortunately UK funding was terminated some years ago and the telescope was taken over by the University of Hawaii I believe. The telescope was closed down a while ago, together with two others on Mauna Kea, possibly in an attempt to appease the protesters objecting to the construction of the Thirty Meter Telescope. Mauna Kea is considered to be a sacred place by native Hawaiians. John

The 29th August marks the anniversary of the passing of David Sinden former chief optician at Grubb Parsons, Newcastle. Although not large by today's standards David worked on many telescopes which at the time represented the cutting edge of telescope technology. These included the Anglo Australian Telescope (now renamed the Australian Astronomical Telescope), The Isaac Newton Telescope and the United Kingdom Infra-red Telescope (UKIRT). After leaving Grubb Parsons David went on to establish his own very successful business, the Sinden Optical company. Despite his pedigree David used to refer to himself as an amateur and always gave feely of his time and knowledge. Meeting David was an unforgettable experience such was the nature of the man. The attached image shows David (wearing a waistcoat) with the 72 inch mirror for the Helwan Telescope Egypt. To find out more about David and Grubb Parsons please have a look at my website: https://sites.google.com/site/grubbparsons/home Thanks for looking. John

Just to give an alternative take have a look at Mel Bartels's notes.... http://www.bbastrodesigns.com/JoyOfMirrorMaking/Parabolizing.html ​ John

Stub M, Those old blanks had very poor thermal properties, the thickness of the blank resulted in a long cool down time which was a problem. Even the small mirrors in amateur telescopes can suffer from the same problem hence the need for cooling fans. The development of new types of glass/ceramics with far superior thermal properties made the 'old' mirrors redundant. All professional observatory telescopes now have mirrors which are either thin or have a 'honeycomb' type structure to aid in rapid equilibrium between the glass and the air. John

Good call RAC John

I think the confusion has arisen because an interferogram can look similar to a ronchigram in that both have a series of bands running vertically across the mirror. A spherical mirror will show straight bands in either test. For a paraboloidal mirror things get a little more complicated. I hope this helps make things a little clearer. John

Ian, you are confusing the Ronchi test with an Interference test. There is no cancelling of light due to interference in the Ronchi test which is essentially a variation of the Foucault knife edge test. The bands seen in the test are giving information about the local slope on the surface of the glass, as is the case with the knife edge test. A spherical mirror shows straight bands with the Ronchi test and greys out evenly (nulls) with a knife edge test. Unfortunately the shape of the finished mirror is not spherical, it is paraboloidal. John

Oh Yezzz John