Better 3D Printing Through Holography

When most of us think about 3D printing, we usually think about a machine that melts plastic filament and extrudes it through a nozzle. But we all know that there are other technologies out there that range from cutting and laminating paper, to printing with molten metal or glass. Many of those are out of range for the common hacker. Probably the second most common method uses photo resin and some light source to build the layers in the resin. Researchers at Lawrence Livermore National Laboratory (LLNL) and several universities are experimenting with a new technique that exposes photo resin using three lasers, printing an entire object at one time. You can see a cube formed using the technique in the video below.

In all fairness, the process really isn’t holography but LLNL refers to it as “hologram-like.” In fact, it appears the lasers project more like an oblique projection (you know, like in drafting) which is considerably simpler. Simple enough, that we can’t help but wonder if the hacker community couldn’t develop machines based on this principle. The key would be arranging for the resin to only cure where laser light overlaps.

In addition to being fast, the researchers note that because the object forms all at once, it doesn’t have problems associated with layering or being formed in a particular direction. The technique has printed beams, planes, struts at arbitrary angles, lattices and complex curved objects.

There’s clearly work to be done to improve the process. Complex objects would require lots of lasers. There is also a fine balance between underexposing the resin and overexposing it. Better laser controls and better photo resins would help, but there are doubtlessly new techniques to develop, as well.

The hacker community contributed a lot to the growth of fused deposition 3D printing. It would be great to see hacker innovation on this technology as well.

LLNL does a lot of interesting things with 3D printing. If you want to look at what conventional laser resin printers can do, check out the Midwest RepRap Festival.

22 thoughts on “Better 3D Printing Through Holography

    1. How well can it scale? These models are very tiny. Resolution also seems awfully low. This certainly has promise but it also has a long way to go and a number of potential issues that may not be clear at first glance.

        1. The first iteration is always going to be an experimental phase. I am more curious about the physical limits of how one can physically scale this because clearly it can be refined but at some point, how do you physically add more lasers to scale it? Or deal with interference issues. That sort of thing.

      1. I wonder when the word hologram lost all meaning and became a word for a magic 3d ghost image? Nobody seems to associate it with microscopic interference patterns on a photographic plate.

  1. Anything with internal structures is going to be an issue as it’s going to need to be exposed first due to the change in refractive index of anything formed outside of it, as well as not being in a sealed void, unless of course, you just want to make an instant photo resin container that has to be kept in the dark to prevent it from solidifying.
    Useful for simple open structures to be quickly formed, but still quite limited. The ability to converge the lasers in complex 3d image forms coupled with the software to translate the refractive index of external structures to collimate the internal image projection is needed. I believe the point is single shot production. Adding more exposure steps helps to a point, but it’s not long before you might as well have just layered it in the first place.

  2. “There’s clearly work to be done to improve the process. Complex objects would require lots of lasers. There is also a fine balance between underexposing the resin and overexposing it. Better laser controls and better photo resins would help, but there are doubtlessly new techniques to develop, as well.”

    How about resin costs?

    1. No, not even close. To make those cubes they usually use a 532nm Q-switched laser. Where the beam is focused in the glass block it vaporizes the glass and creates a tiny little bubble. Move the block to the next point and fire the laser, step and repeat.

      1. Although you could use a similar principle, the focal point of UV light would harden faster than the unfocused beam. So you could use a similar method of controlling the focal point to harden resin, but I don’t know that the hardened resin would actually stay where you want it.

        1. You could use a thicker resin, but you would have to start with a really wide beam, much wider than most lasers out out with normal apertures, to get a diffuse beam without hardening it. Thats why they are using intersecting beams here.

  3. I’d like to see how it handles corners once somebody scales it up. FDM printers get that bulge on corners because the print head has to slow down slightly to avoid trashing yet the plastic flows at the same rate. Laser projectors have to slow down as well but you can continuously adjust the power so that the corners get the same amount of light as the straights, so they can cure the same.

  4. Dr Elizabeth Downing showed this kind of appraoch at Siggraph Emerging Tech in (I think) 1996.
    It was a volumetric display using intersecting IR lasers into a doped volume which glowed visible where scanning XY beams intersected in the volume and quantum effects meant the doped material just had enough energy to glow.
    Pretty amazing in 1996.
    I think the company still exists:
    http://www.3dtl.com/page9.php
    I really hope they’re working on this….

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s