This morning, a new 3D printing company (Carbon3D) won the marketing lottery, by appearing in a story in the Washington Post, and then being featured pretty much everywhere possible online. They were tweeted hundreds of times.
While I trust that they really are doing something different, the overall technology isn’t new…it is a variation on stereolithography, which predates fused deposition modeling (what most library 3d printers are using) as a technology. It’s not even the first consumer level stereolithographic printer! The Form 1 (http://formlabs.com/products/
In order to figure out what it was that they were doing differently, I had to read their paper that was published yesterday in Science. Unsurprisingly, WaPo got a lot of the tech wrong, or at the very least wrote it in such a way that it is very confused. Take this section:
“To create an object, CLIP projects specific bursts of light and oxygen. Light hardens the resin, and oxygen keeps it from hardening. By controlling light and oxygen exposure in tandem, intricate shapes and latices can be made in one piece instead of the many layers of material that usually make up a 3D printed object.”
“Bursts of oxygen”? You can’t “project” oxygen into a liquid like you can a laser. And “instead of the many layers” is also raising red flags. There may not be distinct layers in the same way as FDM printing, but there must be some form of progressive building.
What is actually going on is that they are, indeed, using a UV projector to selectively harden a photosensitive resin. What is different about their approach is that they are projecting through a membrane that is selectively oxygen permeable, which allows for a “dead zone” of resin that can’t harden (due to the oxygen level), above which the UV sensitivity kicks in and the resin hardens. They call this process “continuous liquid interface production technology” or CLIP.
The paper doesn’t say it outright, but knowing the technology, I’m guessing that their hardening process is a continuous build. Rather than a laser-based traditional resin printer, they are using a projector, which I can imagine is more like a video, continuously painting the surface to be hardened. It would be more like pulling sugar, where the liquid becomes solid as you lengthen it, and there would be no layers per se, but more of a crystalline lattice. This would account for the smoothness of the prints. It is also, to be fair, a complete guess on my part.
This change in the traditional stereolithography process apparently gives them a huge increase in speed, which is the key differentiator here. They appear to be able to print objects very, very fast. It also looks like they have the cash to research and develop it commercially, with both Silver Lake and Sequoia as backers.
So what does this mean for libraries? Honestly, not much for the moment. This particular technology could be very inexpensive to make…or, given the proprietary nature of the membrane and resin, it could be ridiculously expensive. The company hasn’t announced any pricing or even availability, so we really have no idea when it might be available. When it is, I’ll revisit and see what I think for libraries. For now, this is interesting, but just a news item.