Here on Earth, being able to 3D print replacement parts is handy, but rarely necessary. If you’ve got a broken o-ring, printing one out is just saving you a trip to the hardware store. But on the Moon, Mars, or in deep space, that broken component could be the difference between life and death. In such an environment, the ability to print replacement parts on demand promises to be a game changer.
Which is why the recent successful test of a next-generation 3D printer developed by a group of Berkeley researchers is so exciting. During a sub-orbital flight aboard Virgin Galactic’s Unity spaceplane, the SpaceCAL printer was able to rapidly produce four test prints using a unique printing technology known as computed axial lithography (CAL).
NASA already demonstrated that 3D printing in space was possible aboard the International Space Station in a series of tests in 2014. But the printer used for those tests wasn’t far removed technologically from commercial desktop models, in that the objects it produced were built layer-by-layer out of molten plastic.
In comparison, CAL produces a solid object by polymerizing a highly viscous resin within a rotating cylinder. The trick is to virtually rotate the 3D model at the same speed as the cylinder, and to project a 2D representation of it from a fixed view point into the resin. The process is not only faster than traditional 3D printers, but involves fewer moving parts.
Lead researcher [Taylor Waddell] says that SpaceCAL had already performed well on parabolic flights, which provide a reduced-gravity environment for short periods of time, but the longer duration of this flight allowed them to push the machine farther and collect more data.
It’s also an excellent reminder that, while often dismissed as the playthings of the wealthy, sub-orbital spacecraft like those being developed by Virgin Galactic and Blue Origin are capable of hosting real scientific research. As long as your experiment doesn’t need to be in space for more than a few minutes to accomplish its goals, they can offer a ticket to space that’s not only cheaper than a traditional orbital launch, but comes with less red tape attached.
>a highly viscous resin within a rotating cylinder.
>sub-orbital spacecraft (…) are capable of hosting real scientific research
Right, but what then was the point of doing it in zero gravity? If the resin is like gelatin jello anyways to deal with the rotating cylinder and other mechanical vibrations all over the place, what’s gravity got to do with the whole process? Gravity is probably going to be the least of your worries, so what was the point of shooting it up on a sub-orbital rocket? A better test would be to flip the machine upside down on earth and show that it still works despite gravity rather than under microgravity.
It still sounds like they just made up a reason to pay the company out of public money granted through NASA.
I’m okay with them testing it in space; it could be useful given that there are plans for another moon shot in progress. Finding out if it truly works out there? Probably worth it.
As I said below, though, I’m personally much more interested in learning if it’s going to be made available locally. =)
Eventually they would, and they’ve already got the grant to test it on the ISS. The question is, were the sub-orbital tests really necessary or even useful as an example of “real science” performed on sub-orbital flights?
What was the science they gained?
It would seem to be a lot cheaper to test it on a sub-orbital flight and eliminate any zero/low-G issues there. Verse doing the first test on the ISS, then flying it back to earth for modification, then back to the ISS…..
Like what?
Again, gravity is going to cause more issues than no gravity for a 3D printer, so what science do you gain vs. just flipping the whole thing upside down on earth?
It is great you have a hypothesis that it will just work as if it survives being flipped upside down. But that seems like a rather relaxed/unreliable approach to engineering non-trivial devices for zero/low-G environments. Even if it behaves exactly the way you expected to, it is worth testing and proving that at least one.
If you want to talk about what they might be testing for, how about things like unexpected/unintended zero-G capillary action effects, bubble propagation in resin in zero-G, centrifugal effects on resign surface tension in zero-G, etc, etc..
I seems a bit strange to be against testing a device in something close to its operating environment, before deploying it to production (when production deployment costs millions). On the other hand, it feels like Boeing’s approach to things, and it is working out perfectly fine for them.
> (when production deployment costs millions)
The sub-orbital flights cost millions, too. It’s still expensive to send stuff almost to space – hence why the question of what exactly was gained? Mere minutes of weightlessness doesn’t really reveal things like bubble propagation or capillary action, because by the time all your stuff is starting to settle down, you’re already going down.
If you wanted to test things like the resin properties in zero-G, you could send just the resin and/or the printer container up to the ISS so they could see how it behaves – before you send the entire machine up.
>The sub-orbital flights cost millions, too.
No, sub-orbital flights cost $100ks (per seat). So say $2500 to $5000 per kg. Compared to $75000 to $87000 per kg for an ISS trip.
> by the time all your stuff is starting to settle down, you’re already going down
Well, they claim to have finished a print during the flight. So it was long enough to test that the machine works without malfunctioning. Even a short zero-G test is better than not testing it at all.
> send just the resin and/or the printer container up to the ISS
But you don’t get to verify if the machine works that way. There is going to be so much more to this than “does the container leak”.
I do agree that a sub-orbital test isn’t going to be as good as testing it on the ISS. But for the price, having the opportunity to actually run the machine in zero-G and then make changes and test it again, seems like a logical choice (and good value).
This also likely cuts the development time by a significant amount, since it will be easier to get on a sub-orbital flight than waiting for the next ISS flight.
It is also good to see at least some value coming from sub-orbital flights. I think this edge-of-space tourism thing is a waste of resources.
Photo polymer is not stuff you want to spend time breathing. The ISS is a sealed long term environment. Bringing things up there requires extensive design work to make sure what your doing won’t escape your little experiment and contaminate the whole spacecraft. Thats IF you qualiy to take up some of the precious room there. Sending this up on a suborbital craft would be faster, cheaper, and a lot more available to a team.
Sealing it up is something that doesn’t need zero gravity – just flip the thing around and see that it doesn’t leak in any orientation. Why would zero gravity make a difference here?
Wait, really?
Are you serious or are your a troll?
Resin printing utterly depends on gravity 100%, it is the foundation of any resin 3d printer out there. Take gravity out and there is no 3d printing.
Smh you say it’s useless to test… Do you even know how a resin 3dp works? Do you know what these guys are trying to solve?
Depends how highly viscous your resin is. Jelly vs Honey. The former is probably like sending a lead pencil into space to see if it works. The latter is probably like sending a clever pen into space to see if it works.
The device works by rotating the resin while projecting images onto it. If it’s like water, it won’t follow along – it HAS to be more like stiff jello, so you can actually turn the container and not have it wobble around due to external or internal forces and accelerations – because vibrations means your UV projection isn’t hitting the right places.
So if you want to make a resin printer that works without gravity, you actually need to build one that works /regardless/ of gravity – whether it’s upside down or sideways, or being spun around in a circle. If it works like that, it’ll work in the absence of gravity.
For the same reason they took the time to mix concrete in space….to make sure it works the way they think it will.
https://www.space.com/astronauts-first-cement-space-mars-habitat.html
Materials behave in different and unexpected ways in microgravity, and there’s no way to easily test this on Earth. Surely you’re not suggesting that the first time they would test this technology in space is when it’s actually needed on a mission?
If this produces truly viable parts at such an amazing speed, then I wonder if the technology will be available planetside some time soon? From what I can get from the diagram, it looks like it has the potential to be a lot more robust than today’s resin printers, too. That could be interesting.
The “works in space” angle is interesting, but it would seem to me to have much more immediate and widespread potential value on the ground.
Or maybe I’m misunderstanding something fundamental. Who knows?
It’s got the same idea as radiation therapy, where you have a radiation source that is orbiting the patient and moving around constantly. The radiation sweeps across the healthy tissue quickly, while the tumor at the focal point is receiving it all the time, so the healthy tissue gets less radiation than the tumor. The tumor dies while the healthy tissue survives.
Here the same principle is used by rotating the resin, so the parts that are supposed to cure hard are getting more UV exposure than the parts that aren’t supposed to cure. For that to happen, you need a fairly stiff resin that you can actually rotate without mixing it. It should be like jello or ballistics gel in consistency, so it won’t wobble around too much as you spin it. As such, it should be perfectly viable on the ground as well, and being so it shouldn’t matter if there is gravity or not, because the resin is supposed to keep in shape by its own.
I wonder if reversing the setup to stationary resin and revolving projector would allow it to work with standard liquid resins.
The viscosity they use is not at all like jello or ballistics gel: more like corn syrup or honey.
And the paper describes a method to optical rotate the image instead of the resin container. They don’t actually do it, though. Nor do they propose rotating the whole projector, which is a completely viable way to do it. After all, ordinary CT scanners rotate a half tonne of machinery at 2-3 revs/second with about the same precision: This is a far simpler task.
>If this produces truly viable parts at such an amazing speed, then I wonder if the technology will be available planetside some time soon?
It is available planetside already, like most 3d printing technologies it will sit in the 5 to 6 figure price range for 20 years until the relevant patents expire and the consumer market gets flooded with hobbyist machines.
Its hard to say if this technique will ever scale up to something significant though. The Readily3d Tomolite only manages 6mm diameter x 12.5mm height. The Xolo Xube does a massive, by comparison, 50 x 70 x 90 mm.
The Xolo technique is an example of a two-colour (but not dual-photon) method, which allows penetration into deeper material.
So, does it have Berkeley style open source?
Sadly it seems they have abandoned the open source hardware part 5 years ago. They strongly suggesting to wait for a new design which hasn’t been published on the CAL-hardware repo
> aboard Virgin Galactic’s Unity spaceplane
by definition “spaceplane” is a plane that goes to _space_, aka above ~100 kilometers. Did Virgin “spaceflight company” ever reach that?
Thats one definition of space.
Yes, a NASA JPL definition https://www.jpl.nasa.gov/edu/teach/activity/how-far-away-is-space/
This reminds me of those string art drawings that are created by overlapping many strings across a circle. One thing that is difficult to do with those is high local contrast, so I’m curious what limitations there are on shapes that can be printed with this method.
But does it work underwater?
Photopolymerization works by ABSORBING the incident ultraviolet light. In normal resins, the light is completely absorbed by about a millimeter: You CAN’T print anything deeper than that, because the light doesn’t penetrate.
This sets a very finite upper size limit on the use of this tomographic approach, unless you use something more exotic like two-photon techniques.
So what’s their secret? Tiny objects? Unusually-low absorption resin? Two-photon illumination? Or are they just ignoring it?
Yep, pretty hard limit on object size, due to absorption.
The original paper ( https://arxiv.org/pdf/1705.05893v1 ) says: “to address the absorption issue, a photoinitiator with lower absorption in the illuminating spectral regime, Irgacure 819 was used. “, but they omit the crucial information of how deep they can cause initiation of cure.
The Irgacure datasheets suggest depths greater than a few millimeters so will see so much attenuation that the surface layers will always get so much light they will always end up as cured material. This sets the maximum chamber diameter to around 1 cm, though they never mention the actual number. The photographs of the printed objects don’t show scale or units, but one object shown with a finger is clearly just a couple of millimeters in size.
So even though the original paper suggests (in the discussion of the number of required projections), that radii of 10 cm could be considered, the resin properties would not work at that size.
I seem to remember similar techniques from other researchers do use a kind of tpi. They have a plane of light illuminated in the liquid, and the 2D projection of the object coming through at another angle.
The original technique, however, relies on “degree of cure.”
From the 2017 paper, “Computed Axial Lithography (CAL): Toward Single Step 3D Printing of Arbitrary Geometries” (Kelly et al):
“Both resins used in this work are activated by absorption of blue light from a DLP projector the photoinitiator. This triggers a reaction which generates free radicals in the (meth)acrylate end groups of the prepolymer which then form covalent crosslinks between polymer chains. As the exposure dose is increased, more radicals are generated and more crosslinks form. As the local density of crosslinks, or degree of cure, increases, the polymer material transitions from a liquid to a solid and increases in stiffness. The liquid-solid transition can be characterized by the gel point, a threshold in the degree of cure above which the material’s storage modulus exceeds its loss modulus. Near this point there exists a threshold in the degree of cure below which material will be washed away in a development step and above which the material remains.”
Thanks for the clickbait. I’m with dude. What the hell does any of this have a need for microgravity. The lack of recusal here is just amazing.
I’m glad there’s still research going on with CAL. I hadn’t seen any updates for a while now.