MIT’s Glass 3D Printer

How hot does your 3D printer’s hot end get? Most low cost printers heat up to 240°C (464°F) at the most because they contain PEEK which starts to get soft if you go much higher. Even a metal hot end with active cooling usually won’t go much higher than 400°C (752°F). Pretty hot, right? [MIT’s] new G3DP printer goes to 1900°F (over 1000°C) and prints optically clear glass.

By changing design and print parameters, G3DP can limit or control light transmission, reflection and refraction. The printer uses a dual heated chamber. The upper chamber acts as a 1900°F kiln while the lower chamber serves to anneal the structures. The print head is an alumina-zircon-silica nozzle.

There’s a patent filed on the process, apparently, and you can read the technical details from [John Klien’s] thesis and in some upcoming publications. Two things we found interesting about the thesis: [John] is an architect and the printer uses GCode.

If you can’t scrounge up any alumina-zircon-silica and a 1900 degree hot end, maybe cutting glass with a CNC is a more achievable goal.

34 thoughts on “MIT’s Glass 3D Printer

      1. They stop the glass by blowing cool air inside the nozzle, which freezes the glass in the tube similiar to how plumbers freeze water pipes when they want to replace a broken section. When the plug melts, it gets going again.

  1. This seems pointless at this time, but I bet I”ll hear about a perfect reason for this to exists soon enough, maybe not related to glass though but something else that is similar in requirement for printing.

    1. With these additive manufacturing techniques you can often embed prefabs of a different material into the matrix. With for example crystal glass, PbO, you can have a lower temperature where the glass becomes fluid enough to work with, so machine parts you embed won’t warp as much from the heat.

      Glass is a relatively cheap material. Together with a solar furnace you might make rather large structures at low cost.

      Perhaps something for constructing a base on the moon.

    1. If you mean the tiny hanging ones, my guess is that they’re PAR-16 lamps with a brass fixture. They lamps are relatively cheap, give off a good field of illumination, and don’t use up too much power (they’re about 50 watts each, with LED versions available). Your local Home Depot (or equivalent) probably has something similar for way more money than anyone should spend, but it’d give you an idea of what’s out there.

      1. Non-toxic alloys
        62.5% Bismuth, 37.5% Tin has a melting temp of 202C
        Fields metal has a melting temp of 62C

        Lots of others with Lead in them are under 300C

        One that can be extruded well may be hard to find though.

        1. Indeed, it must me extrudable. Maybe aluminun or aluminum alloy in a proper environment.
          I keep wondering about that metallic stuff the dentists used to stuff in our rotten teeth in the past days before the evolution of the rosins (and the mindset of our mothers and fathers about the importance of the toothbrush, lol again)

    1. Metal will simply run away like water, because it has a sharp melting point in contrast to glass or thermoplasts where the viscosity decreases slowly with rising temperatures. That is also why you get holes in sheet metal when welding with too much current. As soon as the surface tension can’t hold the liquid metal in place any longer, it just flows away like water

      1. Serious professional welders can weld, with fully melted metal beads, in an overhead position. Gravity is very much not in your favor in that position, but it works anyway. It’s a touchy operation to get right, but I fully expect an extrusion printer for a variety of metals can work.

        1. True. But that works because the close-by metal cools down the weld puddle and the puddle is small enough to hold by surface tension. I’m unsure if the light arc/ plasma supports the effect..

    2. even if the other mentioned points are solvable you at least need a protection against the athmosphere, like the shielding gas in the welding process.
      3D printing with a welding machine is already done, BTW.

    1. there are some fairly interesting applications if it is.
      though total internal reflection might be just an interesting property to tease out of a 3d printer.

      i wonder how small filaments one can work with, glass can be easily pulled into quite durable threads as it is.
      perhaps one could draw out the glass into thinner sections using rollers or something similar.

    2. For the junction between passes to be optically clear I expect the later pass must be hot enough to melt the surface of and properly weld to the earlier pass. This would also seem to be optimal for strength.

  2. The lower chamber anneals the glass *prior* to extruding the glass or is it the bed that’s in the annealing chamber?

    I can’t see it extruding molten glass sticky enough to attach to the previous layer and still be annealed. If the annealing isn’t done correctly, it places internal stresses on the glass that will cause it to fracture over time. I shelled out nearly $600 for a small annealing box for exactly this reason.

  3. When you think about it this isn’t much of an invention because it is basically what many factories do, like lensmakers like nikon and canon,, they extrude glass into a rod – then cut the rod in discs to make lenses. So the leap to moving the nozzle isn’t all that big.

  4. I’d like to see this progress to the point of being able to print in mid-air like this:

    and once that’s possible, electroplating it,
    possibly similar to this:

    Then you could more easily make glass light pipes for solar optical lighting, or at least make the light pipes for the ceiling fixtures. If the UV filters are left off the roof-top solar collectors, the possibility exists that you could have solar-powered cooking on the stove in your kitchen.

    I’ve been waiting for the prices to drop on this for years:

    I’m not affiliated with these companies (one is a dealer), and have not seen any of their products at work, so I’m not even sure their materials can carry a useable amount of light (I’d define that as enough to read a book without anything close to eye-strain) as far as they claim, but I find the concept very interesting.

  5. Essentially this is just coilpotting melted glass rod. The layers look pretty consistent. If they added a small torch like a gtt cricket aimed at the previous layer it would be possible for the prints to be much more optically clear. The idea is to make the surface of the previous layer just molten enough to avoid stress in the glass.

    Looks like I still have a few more years as a lampworker before the machines take over.

  6. This reminded me of my mom that makes and sells glass beads and other small glass trinkets (mostly as jewellry or decoration) as a hobby. Basically all she does is to take glass sticks/rods of various thickness and colour, melts them a little in a flame and transforms the rods into pieces of art. You can check the web for hand made glass beads, amazing what you can do with only a flame, some glass and some creativity… Not something that can be replicated by any 3d printer ever. But the printer probably has other advantages like precise repetition of tasks that manual workers can’t do quite as good.

  7. Puts in 1PB project silica white coated glass microsd card into hp omen 17 2024 8k hdr 1000 self charging self cooling intel i90 apu rtx 5090 hybrid workstation/gaming gpu touchscreen model with self colling parts and 100TB SATA 3 ssd and 16TB NVME.2 ssds with 1GB 2x 512 ddr5 sodimm ram chips that also self cool.

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.