This Vapour Deposition Chamber Isn’t Vapourware

If you are an astronomer with an optical reflecting telescope, the quality of your mirror is one of your most significant concerns. Large observatories will therefore often have on-site vapour deposition plants to revitalise their reflectors by depositing a fresh layer of aluminium upon them. You might think that such a device would be the preserve only of such well-funded sites, but perhaps [Michael Koch]’s work will prove you wrong. He’s created his own vapour deposition system (Google Translate link of the German original) from scratch, and while it might be smaller than the institutional equivalents it is no less effective in its task.

At the heart of it is a stainless steel vacuum vessel with a two stage vacuum pump system to evacuate it. The mirror to be silvered is suspended in the vessel, and a piece of aluminium is suspended over a coil of tungsten wire that his electrically heated to melt it. The molten aluminium is described as “wetting” the tungsten wire in the same manner as we’ll be used to solder working on copper, but in the vacuum it vaporizes and deposits itself upon the mirror. Such a simple description glosses over the impressive work that went into it.

This is a long-running project that isn’t entirely new, but very much worth a look if only for its introduction to this fascinating field. If you are new to vacuum work, how about looking at a Superconference presentation introducing vacuum technology?

Thanks [Paul Bauer] for the tip.

22 thoughts on “This Vapour Deposition Chamber Isn’t Vapourware

  1. I was realizing a while ago, that this is the only missing step in our hackerspace, to being able to walk in with a piece of scrap glass, and walk out with a complete telescope. We can melt the mirror blank in the kilns, figure and polish the surface with traditional means, build the spider and tube and mount, and align all the optics. Only coating the mirror would have to be done elsewhere… unless we build or scavenge one of these. :)

  2. Use copper instead of aluminum (or maybe silver for a slightly lower melting point), print a mask on the glass first, and voilà! Printed circuits on glass!

    Okay, there’s probably an easier method…

  3. he built an evaporator like some old semiconductor fab equipment I’m a little familiar with :)
    The ones I saw always used diffusion pumps rather than turbomolecular pumps, i’d be curious to see how the pump holds up in this application but it seems so low volume that it’s not a big deal.

    Maybe a lucky HaD reader inspired by this article will find a stash of surplus semiconductor fab equipment and snag an evaporator on the cheap and we’ll be reading bout their find :)

      1. Here is the one article I was referring to: https://hackaday.com/2017/06/11/scanning-electron-microscope-adds-to-already-impressive-garage-lab/

        I guess the solar cell method isn’t really deposition exactly… though is an interesting etching using gases and growth method for polymers: https://hackaday.com/2018/01/04/making-solar-cells/

        I haven’t read into quad junction or other systems that I thought for some reason they used electrospray or deposition methods. HHHmmm… not my area of expertise… maybe someone can explain more clearly.

  4. I want to build one of these, not for telescope mirrors but for putting a shiny finish onto plastic parts. Doesn’t have to be super optically fine, just shiny. Would an old Gast rotary vane vacuum pump be good enough for that?

    1. I’m not sure what the specs are on the Gast rotary vane vacuum pump though that type of pump design is used for low pressure (vacuum) vapor deposition… though I recall from the old Scientific American Van de Graff generator making a mercury diffusion pump apparatus, even though would be kind of gross like toxic potentially, can bring the pressure even lower… though I’m thinking there may be viscous fluids like silicone oil or something that would work similar like a vacuum check valve fluid bottle thing. Yeah, I just checked… I was thinking silicone oil now days: https://en.wikipedia.org/wiki/Diffusion_pump

      Olden days: http://www.sciencemadness.org/talk/files.php?pid=91439&aid=2765&usg=AFQjCNGoxCCAPv74u-agOGZKtRj52aprhA&bvm=bv.147448319,d.amc

      Thanks to Steve for saving time looking for the link: https://hackaday.com/2017/02/16/how-a-van-de-graaff-generator-works/

      1. Gast does not make a pump that would be useful, you will need a oil sealed rotary pump backing a turbo, cryo, or diffusion pump to get to the pressures needed for successful evap coating.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.