3D Printing With Visible Light

This 3D printer manages some pretty fantastic resolution, and these are just the early results of [Junior Veloso’s] build. He put together a machine that prints objects in resin that cures in visible light. To print, a thin layer of raw liquid resin self-levels across a printing surface and a DLP-based projector shines light from below, onto the portion to be hardened. The z-axis then pulls that layer up and the next to be printed will become the newest bottom layer. Horizontally the printer yields 1024×768 resolution with a layer thickness as small as 0.01 mm. No wonder he’s turning out this kind of quality.

The model above took 5 hours to print, with eight-second exposure for each layer, and 0.1mm layer thickness. There is lots of good information on his blog, from the early planning, to the finished hardware so take some time to learn about this fascinating project.

Update: Thanks to reader [Nave.notnilc] for pointing out that we’ve seen a chemical 3D printing technique before.

14 thoughts on “3D Printing With Visible Light

  1. This is excellent.

    My chemistry is weak, but I agree with dude. We need to reverse engineer some of these materials so that we can get the prices down. Those of you who want to research this should examine any of the epoxies, coatings and inks which are advertised as being cationic. You know, as in the majority of modern special adhesives. Look for “Cationic polymerization”.

    You can buy specific UV cure stuff from everyone, ie LOCTITE 3335/3340 at $300+ for 250 ml or it’s cheaper chinese (causes cancer in bystanders within a 5 mile radius) equivalents. Look into the inks used by large format printers used for outdoor advertising. I think Gerber makes these, but any UV cure ink is a start. Same for UV cure varnishes in the printing industry.

    Although UV cure reactions are pretty well documented (it’s like 50 years old), hardening by using visible light is a little more difficult.

    If you can hit a DLP projector with UV, you can use almost anything you can get your hands on.
    I would predict that the DLP portion should be considered a consumable, the same way we would consider an inkjet printer cartridge.

    Basically, anything which hardens or undergoes a stable polymer reaction can be used. If you can handle the fumes (nasty, nasty fumes) you can use various super glues. If you can provide vacuum and temperature control, you can use almost any ink.

    Also, I wouldn’t get hung up on positive processes – at larger sizes and lower resolutions, negative reactions should be possible. In addition, you should carefully look at old school PCB manufacturing processes – remember chemical application of photo-resists? It’s the same process. In practice, it would be fairly slow.

    If you want to formulate something, a co-initiator might be helpful – you start with something which is light activated, then dope it with other chemicals to alter the reaction to meet your needs. I am not a chemist, but again look for “UV sensitizer” and photo-sensitizers for help.

    I suspect that bulk 3D printing fluid is already available at any hardware store in the form of some transparent epoxy based paint or varnish…. if you can find something which is not a fast acting self-solvent and can be thinned while retaining these properties.

    One way around the self-solvent issue is to use a film of material on top of a neutral-transport with higher specific gravity/weight/density than the uncured material. Maybe that’s too much physics, but you can use liquid carriers in certain instances rather than building extensive support structure. YMMV.

    This calls for rather old-fashioned science of the “I got 99 coatings but pitch ain’t one” type pioneered by the lost souls who worked for old T.A.Edison. Good luck!

  2. Aren’t many dental resins UV-activated? I’m not sure of any transparent ones, though, and they’re usually pretty viscous, but it would be a good starting point for experiments with dilution (or a spreader bar).

  3. The catch is that most of the suitable compounds are already marked up at high margins. For people who want to see the process explained:


    @Mr. Hacker
    Very true. BTW, HP toner on their newer printers is slowly approaching the cost effectiveness of ink. Just read a report where actual printer cartridge life for newer HP printers is ~60% or less of the stated number of pages.

  4. Im going to skip most of my comments and just reference the comments of another HAD:


    but to add something new to this discussion….
    anyone interested in this DLPrinter Project, might want to check out the Optoma hd20 with native 1920 x 1080 resolution for just under $1k….not a bad upgrade for a few hundred dollars more.

    In consideration of light intensity…1700 lumen @ 1920X1080 of the hd20 vs 2700 lumen @ 1024X768 his ep728 puts out…

    given the same field of operational focus…you get 2.63671875X increase of pixel density…

    which should roughly correspond to the HD20 delivering 66% higher intensity to work-field then the ep728 despite the lower lumen rating of its bulb. So it should not only increase resolution but also decrease layer exposure time.

  5. heh.
    i have a couple of broken DLP projectors here which have power supply problems.
    i think one is a toshiba, and the input circuitry isnt that complicated on the PSU side so if i can locate a schematic or pinout it shouldn’t be too difficult to kludge something.
    (did this once with a 28″ LCD panel, 9 different SMPSs later and I got a picture!)

  6. Hey, Don’t low-tech TEA lasers generate UV? I can’t remember the focus/aiming tricks, but I remember wearing UV protection around these. I believe that more modern versions of these are used for non-contact labeling of stuff at crappy resolutions.

    However, it should be possible to use a TEA or excimer laser in combination with a DLP array and an x/y stage to greatly increase resolution. Maybe this doesn’t offer a lot of advantage over bulb UV sources, but with the right polar coordinate setup, you should be able to quickly calibrate the setup and then get large area + decent resolution at higher speeds.

  7. Nice links guys.

    Looks like DLP+UV is the way to go, although with 405nm lasers getting cheaper with a 600mW unit for less than £45 this might also be doable.

    I am making progress with my PCB etcher which uses many of the same optics as a 3-D printer.

    Please note, UV lasers are extremely eye unsafe, use approved expensive goggles NOT £10 imported rubbish, your eyes will thank you.

    OD6 at least please, we don’t want any HaD members to get eye damage.

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.