I'm at the Mars Desert Research Station in Utah. We finally got a 3d printed approved for use, and we've been using it to do a few things. So far customized finger splints, a hose for an EVA pack (these are only semi-simulated; it's insanely cold here, and they provide the hot air to prevent face insensitivity and helmet fogging), and a little gizmo I did that lets you use a safety razor as a scalpel. If you have questions, I will try to give a cogent answer but our bandwidth is very limited and we have a simulated speed of light delay so I can't answer straight away!
We are using a Cubify thing because they donated it. The first thing we did was remove the horrible inkjet-like cartridge protection thing they have, and set it up for bulk filament.
For reliability, I must recommend the Solidoodle 3 -- mine kept operating through BEING ON FIRE when I was testing the laser cutter attachment.
We're very bandwidth limited right now, but it basically is like a bread lame.
There's a SCAD script for the finger splints and we've been taking the measurements with a medical caliper. This crew is half medical doctors :) I'm just here to do the engineering and some drone/rover piloting. We're also designing a way to take air pictures from Mars where drones won't work.
Why haven't there yet been any projects for milling in space with some sort of vacuum-remelt process for reusing the milled off materials. 3D printing is great and all, but it still doesn't come close to what you can achieve with milling.
That said, this is awesome. I just think that there are pros and cons to both and we shouldn't be focused only on 3D printing.
Milling is great, but not very mass-efficient. Resisting the cutting forces involved in milling anything harder than light foam, requires a lot more structure than simply moving a deposition print head. Machinists use mass as a shorthand to talk about a machine's rigidity, with anything less than half a ton being basically a toy.
Remelting waste is possible, but this sort of high quality manufacturing is really hard, even on earth with gravity to keep material in a mold and hardly any mass or energy constraints. This, and the chip collection system add additional mass, have to be developed way before being put into orbit, and have to work extremely reliably. Even a tiny amount of swarf floating around the interior of a spacecraft could be disastrous.
We aren't focusing only on 3d printing, we're focusing first on 3d printing, because it can work today, with technology we have, on a low mass budget, and without jeopardizing the primary objectives of whatever mission includes it.
When I toured the Belfast cruiser in London, what stuck out to me (the engineer nerd) was the machine shop. Crammed in a tiny space was a marvelously complete shop, where there wasn't much a skilled machinist couldn't make on the spot. This, of course, is critical for a warship out at sea that would need to make repairs or improvise right now.
In a spacecraft, this is obviously even more critical that one be able to make repair parts on the spot. The Apollo 13 crew was pretty lucky they were able to make makeshift repairs to keep going without much of anything to work with. One shouldn't rely on being so lucky again.
The Challenger disaster (discussed recently on HN) comes to mind. If the crew knew about the damaged wing, and had some ability to make new parts on board, they might have been able to rig a repair good enough to get them home.
I was lucky enough to get a guided tour of HMS Belfast from one of her old officers a few years ago. I think that shop could make just about anything, but there were a couple of things which if broken would have crippled the ship. Some piece of steering gear was the main one, IIRC. Just too big to fabricate on board, and not really replaceable without a drydock.
A minor correction: I think you meant Columbia. Challenger was destroyed by a leaking seal in the solid rocket booster while still in the launch phase. Not something fixable while in flight.
Scratching head on what the long term goal of this experiment was... (more 3D printing PR?)
Seems like using this manufacturing approach would be a very tough sell for any real mission.
Only benefit of 3D printing at your destination is the ability to manufacture something that was overlooked, so contingency planning. (yea, yea, someday we'll mine the printable materials on site, right...)
For just about any other item that you know you need, it would be much more weight-effective (the golden measure in launch considerations) to just build the part here on earth, where you can maximize specific density and specific strength using materials that 3D printing can't touch. Plus you aren't lugging around a heavy 3D printer + raw materials.
The article points to the short term goal being just to learn how 3D printers behave in space. They mention printing things in space and then sending them to the ground for analysis.
Long term, I think the goal would be to do metal printing in addition to plastic printing (most likely with separate machines).
3D printed plastic parts can do a lot - they aren't as weak as people make them out to be.
The obvious advantages are that complex things can be made on site with zero labor required (besides occasional assembly of things that can't be printed assembled).
It can be hard to speculate on what a general purpose tool will be most used for in the future, but I do think that there is something on-site printing offers that delivery from earth does not. As we send spacecraft beyond low earth orbit, sending from earth simply stops being an option.
Hell, from what I recall of the movie a 3D printer onboard Apollo 13 could have salvaged the mission entirely. 3D printing isn't just about making things you forgot, but replacing things or making new things. Maybe people get to mars and realize there is an experiment they want to run that needs some little assembly they could print.
I'm not able to predict what a 3D printer would be used for, but you can be damn sure I'd want the ability to fabricate parts locally if I was on a multi-year Mars mission. And that does seem to be our long term goal.
On an orbiting facility, like the space station, you get some other benefits...
* assuming theres also a filament extruder package (more initial weight i know), you can recycle the printing material some number of times, meaning that you can just not ship a bunch of custom things with each subsequent experiment package.
* Materials can be better packed. Weight is an issue, but so is volume. A roll of filament can produce a bunch of parts that no longer need to be designed and manufactured to withstand "unfolding" stresses - you can just make the awkward shape in situ.
* Related to the previous point - some things can be manufactured with much less material - they never need to handle 1g environments, let alone the vibrations and higher g parts of launch. This means less material in a lot of situations.
* I presume (if there isn't now) there will be "rough" launches to carry bulk materials, and "gentle" launches for delicate things and people. If there is a way to do "rough launches" cheaper, (higher g, higher vibration, more risk since there aren't humans or super expensive equipment on board) then you can provide more space on the "gentle" launches for the the delicate equipment. Further - with more space on "gentle launches" you get a better iteration time, you don't have to schedule space/weight for your 3d-printable parts as tightly.
Basically - 3d printing in orbit provides a nice way to separate several logistical concerns, easing some of the expense and engineering needed per experiment in space.
You could save weight of parts needed to be transported to space.
Here's my reasoning: backups / spares
There are an enormous number of backup parts on the ISS – there needs to be for any manned space vehicle – to fix systems that fail. Instead of wasting weight, fuel & money on lifting multiple spares to the ISS, one could 3D print them on demand. Not all spares will be needed simultaneously.
"You wouldn't download a $physicalObject, would you?" has been an annoying meme for quite some time.
Every year, that meme is becoming closer to reality. You can already download small plastic objects. And I look forward to the day when I can torrent a pirated Tesla to my garage :p
Imagine what the film Apollo 13 would have looked like with a fast 3D printer on board. Instead of trying to shoehorn a square CO2 scrubber into a round hole using all sorts of stuff that was never designed for the job, Houston could have sent them a perfectly designed adapter.
As a (Mechanical) engineer, I would respond by saying you turn a bolt with a plastic wrench just like any other wrench, providing the moment of torsion provided by the person turning the wrench does not exceed the shear strength of the wrench material.
Using a socket wrench rather than, say, a crescent wrench means there's more nut in contact with the wrench at any given moment. So the wrench doesn't have to be quite as strong, since the torque is distributed over a larger surface area and thus a larger volume of wrench is available to absorb it.
We use some where I work. They are useful in niche applications, such as our one-off research lab equipment. They wear out slowly. They work best where we have 5 different common orifices for fluid that need to be merged into one. When they wear out, we order new ones.
We pay a tech who knows CAD and a fab house rather than a machine shop. A machine shop would probably cost a lot more. Although now that we have the design down pat, we wouldn't waste as much money on prototypes.
Our design includes 2" diameter threaded ends and it holds pretty firmly although the threads do wear out and tear off the tube with vibration and heat changes.
We've used 3D printed plastic parts in surprisingly (to us, anyway) rigorous applications. We've used printed PLA parts as prototype bearing holders on seriously high speed rotating equipment. Strength is usually close to fiber reinforced Delrin or similar plastics.
3D printing let us try and tune successive iterations for best performance far more rapidly than we could have done with classical subtractive machining. OTOH, our 3D printers care about the room temperature and HVAC breeze, whereas a lathe or mill wouldn't notice. Tradeoffs....
I recall reading about plans to manufacture fuel on the surface of mars for a return trip. I wonder if the goal now is to manufacture fuel and raw 3d printing materials? Seems like it takes some of the pressure off the initial trip loadout.
I hadn't thought about it like that before, but 3D printers are very much the first step of a replicator.
It would be nothing to throw some speech recognition into the loop - not that you need to - and basically say "I want a 16mm wrench" or "I want a 16 oz cup" and come back in 1 hour to find it exists.
