The Ins And Outs Of Casting Lenses From Epoxy

Epoxy lenses

If you need a lens for a project, chances are pretty good that you pick up a catalog or look up an optics vendor online and just order something. Practical, no doubt, but pretty unsporting, especially when it’s possible to cast custom lenses at home using silicone molds and epoxy resins.

Possible, but not exactly easy, as [Zachary Tong] relates. His journey into custom DIY optics began while looking for ways to make copies of existing mirrors using carbon fiber and resin, using the technique of replication molding. While playing with that, he realized that an inexpensive glass or plastic lens could stand in for the precision-machined metal mandrel which is usually used in this technique. Pretty soon he was using silicone rubber to make two-piece, high-quality molds of lenses, good enough to try a few casting shots with epoxy resin. [Zach] ran into a few problems along the way, like proper resin selection, temperature control, mold release agent compatibility, and even dealing with shrinkage in both the mold material and the resin. But he’s had some pretty good results, which he shares in the video below.

[Zach] is clear that this isn’t really a tutorial, but rather a summary of the highs and lows he experienced while he was working on these casting methods. It’s not his first time casting lenses, of course, and we doubt it’ll be his last — something tells us he won’t be able to resist trying this all-liquid lens casting method in his lab.

10 thoughts on “The Ins And Outs Of Casting Lenses From Epoxy

  1. No doubt that size limit can be pushed a little further,especially with the right choice of liquid. Perhaps two liquids could be used, one with a density and one with a higher density than the lens medium or even a third to add a central bulge. I wonder how this compares to pulling a vacuum behind a sheet of plastic, in a frame and using the resulting concave surface as a mould.

    1. Curves made from a pressure difference like that are spherical rather than parabolic, and make poor optical lenses. One way to make a natural parabola is to spin a basin of liquid, this causes the liquid to sort of hug the sides of the basin and it happens to make a parabola. This is still challenging for lens making as things like resin still warp while setting, and you have to spin the basin at a constant speed. The accuracy of the lens has to be better than the wavelength of the light you want to focus, which means you need precision in the nm range.

  2. Although not a precision technique, one trick that might be worthwhile for salvaging a foggy plastic lens is flame-polishing. This technique is well known for glass (albeit, not necessarily for glass lenses), but works — sometimes — for plastics as well. The plastic must be a thermoplastic, which I’m not sure epoxy is. The technique is to run a hot flame over the foggy plastic surface quickly. MAPP gas, through a swirl torch, works well. It takes practice, and there’s no certainty that the optical characteristics won’t be altered. Hence, it’s good for polishing fogged safety glasses (plano) but questionable for polishing lenses.

    1. If the imperfections are all on the surface, you can mount the lens on a slow turning lathe – and add another layer of epoxy resin.
      The turning lathe stops the epoxy dripping off (akin to a roast pig cooking on a spit). You don’t need a diamond prcision lathe for this level of work.

  3. Turntable, or an old 48V fan as I used for spin coating experiments.
    If anyone else runs into these, they are from old 2G base stations and are really high quality.
    Full speed control, and also can be run from 26 to 48V without much of a problem.
    Also another method would be to locate a cheap turntable intended for use by children and hack that.
    Quite a lot of these had the same basic hardware but terrible quality components on the amplifier side
    so would generally only play their own optimized-for-crappy-hardware discs well.

    Hope this helps!!

  4. One of the “Amateur Telescope Making” volumes spoke of making a mercury mirror (a century ago!) by spinning. If memory serves, vibration was reduced but not eliminated using a soft rubber drive belt. Optical quality was fair. For a good telescope mirror the wavefront error should not exceed about 1/8 wave or about 80 nanometers; today’s serious amateurs lean toward errors a fraction of that size. Getting that kind of surface by spinning a liquid is likely to be a lot more challenging than it might seem at first.

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.