Clever Chemistry Leads To Much Faster 3D Printing

Resin printing, it can be messy but you get really great resolution thanks to the optical nature of curing the sticky goo with light from a projector. Soon it will have a few more notches in its belt to lord over its deposition cousins: speed and lack of layers. A breakthrough in resin printing makes it much faster than ever before and pretty much eliminates layering from the printed structure.

The concept uses an oxygen-permeable layer at the bottom of the resin pool. This inhibits curing, and apparently is the source of the breakthrough. The resin is cured right on the border of this layer and allows for what is described as a continuous growth process rather than a layer-based approach. One of the benefits described is no need for resin to flow in as the part is extracted but we’re skeptical on that claim (the resin still needs to flow from somewhere). Still, for us the need to work with resin which is expensive, possibly messy, and has an expiry (at least when compared to plastic filament) has kept deposition as a contender. The speed increase and claims of strength benefits over layer-based techniques just might be that killer feature.

The technology is coming from a company called Carbon3D. They are branding it CLIP, or Continuous Liquid Interface Production. After the break you can see a video illustration of the concept (which is a bit too simple for our tastes) as well as a TED talk which the company’s CEO, [Joseph Desimone] gave this month. Of course there is also the obligatory time-lapse print demo.

So what do you think: game changer or not, and why do you feel that way? Let us know in the comments.

[Thanks Kendall14 via Scientific American]

48 thoughts on “Clever Chemistry Leads To Much Faster 3D Printing

  1. OK I’ve never personally seen a traditional resin printer that has printed a layer, pulled the whole layer out of the resin to refill the reservoir, reposition the print in the reservoir and then print the next layer. My understanding is that they effectively work as they demonstrate their “CLIP” system functioning sans the oxygen layer. Tho in fairness I have never actually owned a resin printer.

      1. Using the oxygen created dead zone is the game changer. It allows fresh resin to flow under the part, the light to shine through and cure past the dead zone. I hope they make these available as heads for existing CNC machines rather than try to sell the whole machine.

        1. The only reason you need the semi-permeable membrane is to keep the resin from dripping out the bottom.
          What would happen if you flipped the geometry like the peachy used an oxygen rich environment an a lower surface tension resin?
          I release this idea under the creative commons license.

      1. haha the peachy is a 100 dollars what do you expect? It is also a first gen of the drip type, but guess what it is open so anyone can improve upon it. The idea is better for making cheaper printers as it doesn’t require a DLP projector. I wasn’t comparing performance neither was he, somethings like ethics are slightly more important.

        You will realize that someday, when you grow up.

        1. Cheap is worthless if it’s slow. I’ll gladly pay twice as much for a printer that prints in half an hour what another printer would take a full day to do. DLP may be costly now but the price will come down, and curing an entire 2D plane at once is always going to be hundreds of times faster than scanning with a laser.

          1. That depends on the speed and power of your laser.

            The issue of DLP printers is the limited resolution. For larger objects, the projected pixels become larger as well, whereas a scanning laser or light dot can have arbitrary resolution at any projection size, only limited by the tolerances of the scanning mechanics.

            A typical DLP mirror array has 15 micron wide pixels, but at 1:1 magnification your print volume is limited by the size of the chip, which is around 1″ square. If you want to print an object that is 3″ wide, your resolution drops to 45 µm as you increase magnification.

            The surface quality of an object at 20 µm matches that of sandcasting or a roughly sawn surface. That’s why large format 3D printers are forced to use scanning lasers – otherwise the objects would turn out like Minecraft.

      2. We can fix the resolution… that’s actually more simple than you might think. But as an out of the box printer, you’re right. It does however blow the plastic extrusion ones out of the water even on their best days.

    1. I believe the difference is that you still have to wait for the resin to flow back into the previously exposed region (the delaminate-recoat-reposition sequence in peach, as I understand, skips the delaminate-reposition but still has to wait for the current surface to be recoated)
      With this scheme, the resin is already there (it’s in the oxygen infused region – the “dead zone” they call it) and as hardened resin is drawn up, away from the zone, fresh resin is *instantly* drawn away from the oxygen rich region replacing it.

      If the resin used on other printers (not just peachy) had a very low viscosity and could harden very quickly then it could be comparable as the flow back over the freshly exposed surface would make the ‘delaminate-recoat-reposition’ sequence negligible.

      1. Pulling on the resin too fast creates a low pressure that pulls oxygen through the semi-permeable membrane and extends the dead zone up to the object itself, which means you can’t pull it arbitrarily fast.

  2. I like this except the one fatal flaw, a patent. Greedy adolescents need to grow up some time in the next 100 years and realize discoveries should be shared and used by everyone.

    So no, not a game changer, it would be if it was open source and open to improvement whilst promoting competition.

    1. You’re saying the development of new ideas and techniques (even when protected by patents to allow financing of further research etc) don’t promote competition??? Open source is great, but how quickly do open source developments move by comparison to paid for research? And aren’t the majority of open source things developed as a direct reaction to closed source or patented things having already being developed?

        1. How about this:

          Major research takes a lot of money. Some of the outputs of this research are valuable (some are not) and easily copyable. Being able to patent means that you can be a little more confident about making a return on your investment. No patents -> less investment.

          That being said, I feel that there are a lot of crap patents out there (example, APC’s patent for a 3 level NPC leg where they’ve replaced the outer IGBTs with MOSFETs) that should never have been granted.

        2. That’s not what he said at all. Don’t be a troll.
          He said that patenting technology slows innovation, which is a reasonable, if possibly dubious, point nowadays.
          Originally, patents were designed to increase the development rate of public domain technology by providing a short temporary monopoly to the original inventor in exchange for detailed documentation of the invention so that it could easily be used by everyone once the patent expired.
          This has three different effects on innovation.
          The first effect is that return on investment for research is more certain, due to the monopoly rights. This should increase innovation.
          The second effect is that new inventions are mostly kept out of the hands of the public for [patent term], slowing widespread adoption of and further development upon all patented technologies. This should decrease innovation.
          The third effect is that inventions that were previously patented with detailed documentation will never be lost, preventing technological backsliding.

          When patent terms are short relative to the iteration cycle of technological development, in an environment where technology older that [patent term] is useful, and when technology is at risk of permanent disappearance if not centrally documented, patents do more good than harm.
          This was the case when the patent system was designed.

          However, the iteration time of development has only gone down, the useful lifespan of most inventions has massively decreased, and communications and information technologies allow even very small groups of hobbyists to keep technical knowledge alive today.
          We should reexamine the costs and benefits of the patent system, because it obviously could use serious adjustments, and plausibly may not even be beneficial at all anymore.

        3. The whole point of patents is that in exchange of publishing all the relevant parts of the invention for all to see, you get a time limited legal monopoly to it.

          The alternative to patents is that researchers start to keep every invention locked up in a safe, paranoid about ever showing them to anybody lest the Chinese get a wind of it and beat you to the market.

    2. I bet once it or a competitor has fantastic results and others want to use it, one amazing hacker/pirate will get the necessary info from the patented BS and turn it opensource under a new name, and distribute freely. Patents only matter if you’re trying to sell it. Open source is neat because inherently, it’s free or base value. Then there’s this asshole makerbot industries…

    3. may i point out that one is allowed to do whatever he wants in his own home, including building copyrighted machines and tools, as long as one doesn’t use them commercially?

      you can read a patent and copy it to play around with it all day you want. as long as you dont sell it or any of its products

      1. You can build a product by reading patents, but to sell the product you must change enough of the patented pieces to become technically different. Unfortunately, patent holders tend to submit several variations of a device by brainstorming all the different ways to build said device. That’s where the greed comes in.

  3. Not a game changer IMO. There are some inherent negative features of resin that still keep it as only a contender with filament based printing, I don’t care how fast. The resin itself is mostly what I have in mind, I don’t think it’s as broadly diverse as the many kinds of filament available. The fancier mechanism also is a negative. The extra quality is a good selling point, but think along the lines of the Sony Betamax vs VHS. Filament based printers are well-ensconced and “good enough” for many things. The patent bits put the nail in the coffin. Don’t buy it.

    1. The only thing resin seems to really excel at is resolution/detail. In fact, I would go so far as to call that their niche. On the hobbyist level, they are ideally suited to generating high-detail parts either for prototyping things that are to be produced another way– i.e. if you want a decent preview of the finished product before you invest in tooling –or a master from which to make RTV molds and cast copies in much sturdier resins.

      For me, if I could afford a resin printer, one of the primary uses would be making masters for casting my own miniatures and ‘garage kit’ models and figures.

  4. Where should I start?
    Well the first thing I would say is that what people associate with the concept of “3D printing” is FDM machines and more rarely curable resin printers. That is only because they are what consumers are offered thus that’s the word getting around.
    But there are several of other varieties of 3D printers either in existence or being developed over the past several years. Some can make metal parts directly and some biological tissues for instance. So add another method and it’s in the mix, that’s all. Today it seems like it’s more a concept than a method, and the definition is getting more broad over time.
    If it is expensive and closed, then that is two major factors in the way of it becoming popular right out of the gate. Too bad for the short-sightedness I guess. If not, and if it is as fast as they claim, then I can see it becoming popular from that alone. Sounds like it will probably be expensive though and in that case I wouldn’t even be taking a look at it anyway. I’m not their niche, so oh well. I’m fairly happy with what I’ve got. I’m sure the technology will jump sometime in the future and a cool whizbang thing I can afford will come along. Until then I’ll keep flaking my own spear points and have some fun along the way.

  5. This is very cool. Speed is a rate limiting step in prototyping with all 3D printers currently. I would love to have one of these to work the kinks out of a design.

    Never been a fan of resin as a material, but who cares? That’s what molding and casting is for.

    I’m just glad to see an advance in 3D printing that isn’t “make it 5% cheaper than last year”. I hope they do great, and I hope to see more tech advances in 3D printing to follow this one.

  6. It sure seems promising as far as a resin printer can, however, is this actually true? Because it looks exactly like a normal resin printer and those SEM pictures have no scale. There is no actual proof.

    1. I think weight to hold strength at the material union on the stage is going to be the scale limiting factor, unless that is in some way made to provide a mechanical union as the part is extruded.

  7. I have a hard time putting faith into any company that is self proclaimed to be “game changing.” And, I like that they mention the aspirations to use many different materials and the whole shoot high aspect, but it is annoying how the presenter states that these things seem like they will be reachable tomorrow. Different resins haven’t changed that much, and the product still needs to be light curable, which really limits the structural strength of anything that can be made in this process.

    I still think resin printing has its place, but speed isn’t the benefit of resin printing imo. The benefit is the lack of need for support structures.

    1. Even resin needs support structures – it keeps those islands from floating away.

      The attraction of resin is that there are different limits to the speed of the operation. With enough emitters one could have any section printed as fast as the resin could cure – where extrusion printers are rated in terms of volume they can deposit, a resin printer is rated in terms of linear dimension. To go from 1x1x1 cube to 2x2x2 cube increases the time on an extrusion printer by a factor of 8; on a resin printer by a factor of 2.

  8. The cost of DLP is prohibitive and we haven’t seen the price drop at all. I can’t see this going mainstream until the cost for DLP is roughly equivalent to filament.

  9. Cool to see the rapid in rapid prototyping with this machine/technique.

    I suspect it has a problem with large cross-section areas. All of the show-case parts a spindly. The video of the Eiffel Tower build shows it pausing a few times. Can fresh liquid resin get into the middle of large cross sections? Do they have a way for the machine to handle this?

  10. “We should reexamine the costs and benefits of the patent system, because it obviously could use serious adjustments, and plausibly may not even be beneficial at all anymore.” depends on the direction, when a big company has control of the patent then yes, shorten it’s term. if a small company or hobbyist has control then keep it at a long term. so that perhaps a large company will want to buy it from the hobbyist. and if the short term of ownership for a large company will help motivate them to either crap or get off the pot with it.

    1. Too many loopholes, the same as in the copyright system.

      If individual owners have longer patent rights, then companies simply write all their patents to individuals and then make them sign an exclusive contract of cooperation.

  11. Ok, Just Stop Now!

    i’m still building my first FDM printer. It’s taking me a while. I declare a moratorium on further 3d printing development until I can get my printer finished, master using it and start itching for something better.

    i don’t want to fall too far behind before I even start!

    j/k… mostly

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.