3D printing

[Jonfliesgoats]

Can you use a 3D printer to print an assembly that has moving parts?  Can you 3D print an axle, wheel and bearings so that no seams, welds or bolts are required?

[wumpus]

How big a gap are you willing to live with?  My guess is not as big as the 3d printer will need.

3d printers aren't the magical boxes some people think they are.  For this type of thing "some assembly will be required".  Also expect that for parts that have to move against one another, CNC devices will routinely produce superior products over 3d printers.

- insert rant about how "3d printing" exists pretty much for low volume applications.  Try to use 3d printing on a larger scale and a competitor that uses vacuum molding will crush you on costs and manufacturing tolerances every time.

[Steel]

I know that Koenigsegg produce some of their car part by 3D printing moving parts inside other others (if I can find a link to something I'll post it here).

[Jonfliesgoats]

I was just curious, really.  There is a lot of hype about additive manufacturing in everything from the US government's Global Trends document to fluff pieces in magazines.  Sort of like autonomous UAVs, it's hard to sift the facts from the hype.

Can your use a 3D printer to print simple sensors, like thermocouples or strain gauges into a maeterial along with associated circuitry?

The GEnX engines on the 747-8 and 787 use printed parts as well.

[Elthy]

5 minutes ago, Jonfliesgoats said:

Can your use a 3D printer to print simple sensors, like thermocouples or strain gauges into a maeterial along with associated circuitry?

It boils down to what you consider a 3D printer to be. Im sure its somehow possible to build a 3D printer that can somehow print e.g. a bimetal thermometer, but it will be way more expensive that using massproduced parts. There are 2 fields where 3D printing makes sense:

-Prototyping/low volume production
-Complicated (e.g. hollow) shapes

For everything else its often possible to use 3D printing, but way cheaper to use something else.

[Jonfliesgoats]

Could you put a 3D printer someplace far away with materials and produce instruments on site?  Say we fly a printer with some reservoirs of different materials to a distant body.  Could we get some utility by having the ability to make more thermometers on wheels or barometers on wings via a printing system?

Could we use this manufacturing tech to provide more freedom to research things as they catch our interest?

Of course, I recognize that with today's technology it is more effective to send lots of existing sensors to a location rather than waste mass shipping a small factory the moon or elsewhere.  I am just thinking about possibilities decades in the future.

Can this type of thing be used for more utilitarian purposes?

http://the3doodler.com/

Also, I want a 3doodler.

[Elthy]

I somehow doubt you can build a omnipotent 3D printer in the close future, but there is a 3D printer aboard the ISS, they test to use it for manufacturing replacement parts.

Also it could e.g. be more mass effective to launch a 3D printer and a few complicated components (e.g. sensors) to a mars colony, they could build the structural parts from local materials.

[Jonfliesgoats]

Interesting.  So you could use local materials with some refinement in a printer?  Would that require some sort of small smelter or something?  Being able to shovel local materials into a hopper and come out with 3D printer structures sounds amazing.

Edited December 30, 2016 by Jonfliesgoats

[Kerbart]

Surface finishing requirements will likely be an issue. For infrequently rotating parts with low loads (a lid on a box for instance) 3D printing will do just fine. An axle/bearing assembly for a combustion engine might be a different story. Jet engines using 3D printed parts sounds cool... but what parts? Moving parts? Fuel lines? Wire harnesses? Are they placed straight from the printer, or require post-print processing (surface finishing, hardening, etc)?

3D Printing offers opportunities in manufacturing that have never existed before, but as with every manufacturing method it will come with its own weaknesses that will counter the possibilities. The challenge in engineering will be to overcome those problems.

[Jonfliesgoats]

You know, as costs come down and materials improve, it will be interesting to see what manufacturing for the masses brings.  

[Shpaget]

Yes, 3D printers can make nifty parts, but they are still mostly toys.

Sure, they can make an intricate thing like this:

 

but mechanical properties of it are very poor. After all, it's plastic. There are printers that use metal, but the material of the finished product is still inferior to other cheaper manufacturing methods.

3D printing is and, I suspect, always will be more expensive and of inferior quality than countless methods of mass production. That being said, they are great for one offs, prototypes, concepts etc. It's great to be able to design a product and print out its enclosure in matter of hours for a low cost, adjust the design and print it again and again until you have ironed out all the issues.

With standard manufacturing methods, such as injection moulding, producing multiple tools for testing and prototyping would be prohibitively expensive, however, once you 3D print your final design and decide it's ready for production, you will go back to injection moulding.

As for printing sensors and other devices you mention, no that will not happen. Not only are 3D printers incapable of such intricacies and material variation, the instruments need calibration, testing, verification etc... Even if you could print bimetallic switches, they would be inconsistent and useless.

[kerbiloid]

3 hours ago, Jonfliesgoats said:

Can you use a 3D printer to print an assembly that has moving parts?  Can you 3D print an axle, wheel and bearings so that no seams, welds or bolts are required?

Yes.

3 hours ago, wumpus said:

How big a gap are you willing to live with?

Gaps.

Spoiler

 

[Steel]

50 minutes ago, Shpaget said:

Yes, 3D printers can make nifty parts, but they are still mostly toys.

Sure, they can make an intricate thing like this:

 

but mechanical properties of it are very poor. After all, it's plastic. There are printers that use metal, but the material of the finished product is still inferior to other cheaper manufacturing methods...

Christian von Koenigsegg disagrees with you, he 3D prints turbochargers for his hypercars:

http://www.turbodynamics.co.uk/media/blog/3d-printed-turbocharger

 

[GluttonyReaper]

3 hours ago, Jonfliesgoats said:

So you could use local materials with some refinement in a printer?

Not just that - it saves on the amount of spare parts and simple tools you have to carry, in theory. Instead of carrying a box-full for each of 10 types of screw, say, you can just take a 3-D printer and a single box of metal/plastic to be converted as needed. There's probably practical issues, and I'm not sure how far you need to take it to be worth it, but it looks good on paper

 

[Shpaget]

22 minutes ago, Steel said:

Christian von Koenigsegg disagrees with you, he 3D prints turbochargers for his hypercars:

http://www.turbodynamics.co.uk/media/blog/3d-printed-turbocharger

 

He prints casings for turbochargers

 

Anyway, this is what it takes to make an injection moulded item.

 

I'd say there has to be a very good reason they go through all that effort - superior quality.

The video is in German, but the picture should be enough.

Edited December 30, 2016 by Shpaget

[Steel]

11 minutes ago, Shpaget said:

He prints casings for turbochargers

 

Not just the casing, the turbo housing is 3D printed with all moving parts that control the variable geometry printed within

https://www.youtube.com/watch?v=DNedUZxP8NU&t=19s

[Bill Phil]

http://www.3ders.org/articles/20161101-ge-unveils-3d-printed-atp-engine-more-additive-parts-than-any-engine-in-aviation-history.html

http://www.geaviation.com/press/business_general/bus_20161031a.html

Granted, it's definitely not a desktop printer.... But still, 5% weight reduction and 1% increase in SFC is a lot when you consider the total fuel consumed and the weight savings. Still a demonstrator, though.

Then there's the 3d printed Cobra. That would be fun. Then there's the printed micro jet engine. Then there's 3d printed food. Then there's 3d printed clothes. Then there's 3d printed furniture. Then there's 3d printed car parts. Then there's 3d printed rocket engine parts. Then 3d printed buildings. Yeah. Buildings.

13 minutes ago, Shpaget said:

I'd say there has to be a very good reason they go through all that effort - superior quality.

I wonder how long that will last... Considering we can already 3d print things at microscopic levels (not economically, of course). With some polishing you could make it better, and given the fact that time always marches on, the world will change thanks to 3d printing. But, more importantly, it has changed.

http://web.ornl.gov/sci/manufacturing/media/news/detroit-show/

Granted, it's just the body and part of the frame. Of course, it was polished and smoothed afterwards, but it was still 3d printed.

3d printing will likely start by making molds and tooling on a much faster basis, reducing the cost of retooling factories for large companies as well as starting new ones. It won't take over mass production that fast, though. Rather, it will help mass production and reduce the cost of setting it up.

Even so, technology is changing all the time. 3d printers can print wood. Not a wood substitute, but actual wood! It's pretty crazy... But 3d printing is going so fast that some of the hype might be justified. Not all of it, certainly, but a good chunk.

[K^2]

3D printing is a very broad category. Most people think of something like thermoplastic extrusion-based printing. It's cheap, widely available, but has very significant limitations. You can't print something that isn't connected to everything else.

But there are many, many other methods of additive manufacturing that fall under the 3D printing umbrella. Direct Metal Laser Sintering, for example, can produce parts with precision of a few microns and with no requirement for any sort of connection. You just have to clear gaps of loose metal powder when you're done. In addition, it can work with titanium alloys. People have used these techniques to print everything from high quality firearms, to turbines, to actual rocket engines.

So in the broadest sense, the answer is certainly yes. You can absolutely print parts with gears, bearings, and all manner of other moving parts. But you aren't going to do that with a printer you pick up on Amazon for a few hundred bucks. It takes equipment that costs in the hundreds of thousands of dollar range, usually custom-built for your particular needs.

[Nuke]

no, at least not with a cheap printer. when i want to build an assembly i usually print one piece at a time, and they my big ol chest of bearings, gears, rods, and screws comes out (many of which were salvaged from junk).

like heres a failed assemby. it failed partially to a design flaw (lack of endstops to keep the shaft from sliding in the berings, and a few bad angles and mis-measurements) and partially to the fact that the plan involved re-coiling the motor, and it turned out it ran too hot for its own good (melted the brush and commutator assembly). its also a fine example of why you dont eyeball things in engineering.

and ive been able to make ducted fans that work quite well:

and this is just a 2 part assembly more or less (the large one had a tail cone as a 3rd part), a hole big enough for a motor on the hosing and a hole in the rotor that accepts the motor shaft (i should have probibly included holes for set screws as well since some of these like to eject their rotor when they get warm).

anyway it takes a good pair of calipers and knowing your machine's tolerances and settings. i know on my machine dimensions less than 0.4mm on x and y, do not exist as far as the machine is concerned. i can set the z as low as 0.1mm so vertical tolerances are a little better. if i need to make an 8mm hole in the xy plane, i usually make it 8.4mm, and i usually still get a snug fit with an 8mm part. all printers are different so ymwv.

 

[Kerbart]

6 hours ago, GluttonyReaper said:

Not just that - it saves on the amount of spare parts and simple tools you have to carry, in theory. Instead of carrying a box-full for each of 10 types of screw, say, you can just take a 3-D printer and a single box of metal/plastic to be converted as needed. There's probably practical issues, and I'm not sure how far you need to take it to be worth it, but it looks good on paper

 

In theory you are right, but screws are about the worst example one can give. Aside from the fact that they're a generic fastening solution (and the whole point of 3D printing is that you have custom parts), strength with threaded fasteners is always an issue (hence 8.8 being the practical standard and not the much weaker 4.6 grade, and even 4.6 would be much, much stronger than anything 3D printed), and you can't really compensate by making them bigger (as this would blow up the size of your design).

On the other hand, tools are a great example of something that can be 3D printed. Instead of taking an entire toolbox into orbit you only have to carry a box of granulate and print the tools on-demand (and maybe even recycle them after use).

6 hours ago, Bill Phil said:

Of course, it was polished and smoothed afterwards, but it was still 3d printed.

But that absolutely nullifies what the OP had in mind: “Can you use a 3D printer to print an assembly that has moving parts?” because now you're back to manufacturing individual parts (as they need to be polished and smoothed) with all limitations that come with it (including the agonizing sentence that every engineering student has to go through: that's really nice, but how are you going to put it together?).

The real benefit would be the ability to fit a bearing somewhere on an axle while it's printed, so the axle can be bigger on both sides of the bearing. That will require extreme high resolution printing. Or perhaps some chemical polishing agent that gets printed in between the rotating parts, and that later gets flushed out and replaced by oil? Once that problem is solved 3D printing will really take off, as the weaker material can be offset by designs that are simply impossible today.

[Jonfliesgoats]

Seems like the technology is rapidly evolving at this point.  Printing a motor or vehicle around an instrumentation package seems like a maybe.

I read here and elsewhere that we can now print wood,  What about printing more energetic fuels?  Could you  print a tailor-made SRB with a rough thrust schedule based on the geometry of its fuel?  A pyrotechnic nerd could make some awesome fireworks.

I ask this because a gradient of materials is used on the nozzles of my engines.  While the printing process between a fuel and an expensive metal would be significantly different, it seems that there could be some amazing possibilities.

A package of instruments that can print a tailored solid fuel rocket could send simple sensors to a broad range of locations for research.

[Shpaget]

11 hours ago, Steel said:

Not just the casing, the turbo housing is 3D printed with all moving parts that control the variable geometry printed within

https://www.youtube.com/watch?v=DNedUZxP8NU&t=19s

You can clearly see the layers on the casing and the bolt for what he calls the louvres. That abysmal surface finish is never going to be good enough for bearings in a turbocharger and I'm quite doubtful it can produce a good turbine.

11 hours ago, Bill Phil said:

I wonder how long that will last... Considering we can already 3d print things at microscopic levels (not economically, of course). With some polishing you could make it better, and given the fact that time always marches on, the world will change thanks to 3d printing. But, more importantly, it has changed.

For mass production? It is likely not going to change. Injection moulding is exceptionally cheap once you have the tools and produces great results, so for large production runs it is the way to go. It is so popular it is everywhere. Every product that contains something plastic is likely to have injection moulded parts in it.

[Bill Phil]

14 hours ago, Kerbart said:

But that absolutely nullifies what the OP had in mind: “Can you use a 3D printer to print an assembly that has moving parts?” because now you're back to manufacturing individual parts (as they need to be polished and smoothed) with all limitations that come with it (including the agonizing sentence that every engineering student has to go through: that's really nice, but how are you going to put it together?).

You can print moving parts (and polishing afterwards doesn't nullify that at all). The moving parts, for now, are at pretty low quality.

9 hours ago, Shpaget said:

For mass production? It is likely not going to change. Injection moulding is exceptionally cheap once you have the tools and produces great results, so for large production runs it is the way to go. It is so popular it is everywhere. Every product that contains something plastic is likely to have injection moulded parts in it.

For mass production you would have to have a bunch of printers. But that would let you print on demand, or have factories on a more local basis. And then it wouldn't really be mass production...

And what's hilarious is that 3d printing will likely make injection molding even cheaper. The technologies will work together before one overtakes the other.

Tooling usually takes a while to make. Retooling factories is a very expensive process. With 3d printing that can be much less expensive and time consuming. Think just a matter of days for new tools.

Also, that Cobra was made in a matter of weeks...

[Kerbart]

3 hours ago, Bill Phil said:

You can print moving parts (and polishing afterwards doesn't nullify that at all).

Polishing afterwards does nullify it. What's the point of printing everything constructed together if you have to take it apart, polish it, and assemble it again? It's easier to print the parts separate, in that case. But than you cannot create "impossible" assemblies that 3D printing makes possible.

[DBowman]

On 12/31/2016 at 3:42 PM, Kerbart said:

the whole point of 3D printing is that you have custom parts

yes, here - but on a long space mission where you want a 99.999% chance to not run out of critical parts ( = loss of crew ) then the mass to ensure you can make the parts as they fail is going to be less than the mass of taking everything that might fail. I mean if 'odds are' that 5 parts will fail and there are 1000 different parts that could fail then being 'sure' all will be well means taking 5000 parts or a printer and 5 parts worth of 'print gunk'. You make a good point re fastener strength etc. Clearly also designing for spare minimisation is important ( Apollo 13 taking square and round LiOH canisters being the example of 'what not to do' ). 

 

On 12/31/2016 at 4:00 PM, Jonfliesgoats said:

What about printing more energetic fuels?

Oh yes! As well as 'continuously variable' SRBs one could do 'digital micro thrusters' - an array of little SRB 'wells' - ignite as many as you need for a articular manoeuvre (maybe they just go up like a box of matches).

Also this kind of thing is pretty interesting, print it flat and origami it into a bot or vessel. Photo lithography is kind of 3D printing on a nano scale - in micro gravity & hard vacuum maybe you can do fun stuff at large scale; vapour deposit printing wiring/structure, UV curing polymers, layer after layer, then folding it up into something.

 

22 hours ago, Shpaget said:

Injection moulding is exceptionally cheap once you have the tools and produces great results, so for large production runs it is the way to go. It is so popular it is everywhere. Every product that contains something plastic is likely to have injection moulded parts in it.

Granted, but 'large production runs' are not the 'long space mission' use case - take 5000 parts vs mass of 5 parts of gunk.