Live data from Hacker News

Carbon3D – 3D Printing via Continuous Liquid Interface Production

carbon3d.com

21–30 of 112 posts

Re: Carbon3D – 3D Printing via Continuous Liquid Interface Production

#21
post #10

The site is so cool that it just screams "fake". Cool pictures, not much copy, no details. The big difference here is that they're doing photopolymerization at the bottom of the tank, rather than at the top like everybody else. This requires a transparent material that passes oxygen on the bottom of the tank, so the action takes place on the surface of that membrane. They're vague about the details. How long does the…

> The site is so cool that it just screams "fake".

They do have a video of the actual product at the bottom, which is just plain cool.

Re: Carbon3D – 3D Printing via Continuous Liquid Interface Production

#23
post #6

Unless I'm missing something, this is not exactly a new technology. http://makezine.com/2014/07/26/droplit-the-low-cost-diy-resi... http://www.instructables.com/id/DIY-high-resolution-3D-DLP-p... http://www.3ders.org/articles/20120911-a-list-of-diy-high-re...

You are. Using UV to cure photo-sensitive resin isn't the innovation, it's using the oxygen permeable membrane.

If you look at the Form 1 printer or the B9Creator, there's a mechanical step between every layer where it needs to actuate the platform in order to loosen the resin from the projection window, so it can build the next layer. In the Form 1, it peels the print off, and the B9Creator slides a window. (look for videos on youtube.)

In both instances, the amount of time spent actuating the mechanical part adds up, and results in a significant amount of time spent in the print actuating the plate. What the oxygen membrane allows us to do is to skip that step between every layer, and simply keep shining a continuous changing image slice of the object as we're pulling the object out of the resin.

Not only does this have the advantage of speeding up prints by orders of magnitude, in materials, the grain of the object influences the type of thing you can built. If you print a stress holding object with the grain orientated in the wrong direction, the part will fail very readily. This way, we have greater design freedom, without worrying about grain direction.

Re: Carbon3D – 3D Printing via Continuous Liquid Interface Production

#24
See the paper in Science released online just now: http://www.sciencemag.org/content/early/2015/03/16/science.a...

It will apparently be the cover story. (Does this count as Sequoia breaking Science's embargo?): https://twitter.com/sequoia/status/577651625545748480

From the abstract: "...feature resolution below 100 micrometers. ... complex solid parts can be drawn out of the resin at rates of hundreds of millimeters per hour."

Key novel feature is the oxygen-permeable, UV-transmitting membrane at the bottom of the tank that creates a thin (down to 20um) inhibited 'dead zone' where the resin can't polymerize. They project the image for the current layer up through the membrane and the dead zone, so the build layer is actually within the tank. This means they can just draw the part up continuously from the top, with no stepping or processing needed after each layer. The thickness of the effective build layer can be controlled by adding a UV light-absorbing dye to the resin, which allows them to optimize for different print speeds.

Here's a relevant patent, issued in 2014, listing the 3 founders of carbon3D (formerly EIPI systems): http://www.freepatentsonline.com/20140361463.pdf

And... founder Joe Desimone also gave a Ted Talk tonight, so soon we'll even get to see a splashy 18-minute long talk about the technology. https://conferences.ted.com/TED2015/program/schedule.php

Re: Carbon3D – 3D Printing via Continuous Liquid Interface Production

#26
post #10

The site is so cool that it just screams "fake". Cool pictures, not much copy, no details. The big difference here is that they're doing photopolymerization at the bottom of the tank, rather than at the top like everybody else. This requires a transparent material that passes oxygen on the bottom of the tank, so the action takes place on the surface of that membrane. They're vague about the details. How long does the…

> The site is so cool that it just screams "fake". They do have a video of the actual product at the bottom, which is just plain cool.

I'm glad you pointed that out, because I completely missed it.

Re: Carbon3D – 3D Printing via Continuous Liquid Interface Production

#27

See the paper in Science released online just now: http://www.sciencemag.org/content/early/2015/03/16/science.a... It will apparently be the cover story. (Does this count as Sequoia breaking Science's embargo?): https://twitter.com/sequoia/status/577651625545748480 From the abstract: "...feature resolution below 100 micrometers. ... complex solid parts can be drawn out of the resin at rates of hundreds of millimeters…

The supplementary materials, including 2 videos showing printing in real-time (keep watching, they slow down), are available without a subscription:

http://www.sciencemag.org/content/suppl/2015/03/16/science.a...

Movie 1: http://www.sciencemag.org/content/suppl/2015/03/16/science.a...

Movie 2: http://www.sciencemag.org/content/suppl/2015/03/16/science.a...

Re: Carbon3D – 3D Printing via Continuous Liquid Interface Production

#28
post #17
post #6

Unless I'm missing something, this is not exactly a new technology. http://makezine.com/2014/07/26/droplit-the-low-cost-diy-resi... http://www.instructables.com/id/DIY-high-resolution-3D-DLP-p... http://www.3ders.org/articles/20120911-a-list-of-diy-high-re...

They're doing the "growing" on a membrane at the bottom of the tank, instead of layer by layer at the top of the tank. That should let them print really quickly and precisely, even compared to the SLA printers you link. The reason it hasn't been possible in the past is that the polymerization process requires oxygen, which has only previously been available from the air over the liquid resin. Their system has some "s…

You've got it backwards about the role of oxygen--it is being used to inhibit polymerization in this case. They are allowing oxygen to dissolve into a thin layer at the bottom of the vat so that the build doesn't stick to the membrane, and they can have a continuous draw.

The true build layer is actually within the liquid, near the bottom, but not on it.

Re: Carbon3D – 3D Printing via Continuous Liquid Interface Production

#29
post #10

The site is so cool that it just screams "fake". Cool pictures, not much copy, no details. The big difference here is that they're doing photopolymerization at the bottom of the tank, rather than at the top like everybody else. This requires a transparent material that passes oxygen on the bottom of the tank, so the action takes place on the surface of that membrane. They're vague about the details. How long does the…

> The site is so cool that it just screams "fake". They do have a video of the actual product at the bottom, which is just plain cool.

I don't know much about 3D printing, but that's the most futuristic video I've seen this decade. It looks like it could just pull a coffee mug or whatever out of nothing and serve it.

Re: Carbon3D – 3D Printing via Continuous Liquid Interface Production

#30

See the paper in Science released online just now: http://www.sciencemag.org/content/early/2015/03/16/science.a... It will apparently be the cover story. (Does this count as Sequoia breaking Science's embargo?): https://twitter.com/sequoia/status/577651625545748480 From the abstract: "...feature resolution below 100 micrometers. ... complex solid parts can be drawn out of the resin at rates of hundreds of millimeters…

Saw the talk, doesn't really add too much.
Post reply on HN