3D Printing: it may change everything, including spacecraft production.

[Moe Grills]

   
Every so often through history, some new revolutionary technological discovery comes along to 'change everything'. (hyperbole, yes; but not by much; a yogi-ism)
Examples:
Gunpowder, steam engines, mass production, electro-magnetism, aircraft, atomic power, transistors, lasers, PC's, internet, etc.
Now, according to an article in the science magazine, NewScientist
(issue: July/30/2011), a recent type of industrial production called 3D printing (still in the experimental stage) promises to change everything
according to Paul Marks (the journalist who wrote the story).

   The gist of this new (possibly revolutionary development) is this:
scan an original item (or CAD something new), digitally 'slicing it up', storing & using the high-volume data obtained for CAM-controled
lasers or electron-guns to fabricate plastic or metal objects
out of powdered plastic or metal that end up as perfect copies of the original.
   Dan Johns, who is using this new method to print strong metal parts
for the Bloodhound SSC rocketcar, claims that the strength of 3D printed titanium metal parts can equal that of the traditional MACHINED METAL.
 
   What this means for rocketbooster and spacecraft production (for parts that are made of plastic or metal) should be obvious.

  This could ALSO mean the beginning of the end for metal lathes, reamers, hydraulic metal and plastic molding, pressing, extrusion devices, and all traditional mechanical metalworking and plastic shaping and molding machines and tools. 
It could also theoretically mean that genuine mass-production of rocketboosters, spacecraft and components can be developed with minimal manpower, and with parts and hardware that can be defect free
(requiring no quality inspection by humans).

Of course, that could still be decades in the future.

[Downix]

Decades?  I'm ordering my 3D printer next week.

[Robotbeat]

Paging user SpaceXULA.

Yeah, people are starting to fabricate rocket engines with this technology. Paul Breed has done it successfully, as has the guy on the rocketmoonlighting.blogspot.com blog.

Here's the rocketmoonlighting guy's test of his Direct Metal Laser Sintering regen rocket motor:



Or maybe this run is better, since you can see the cooling channels:


You can see one of his earlier printed rocket motors:

[Jim]

  This could ALSO mean the beginning of the end for metal lathes, reamers, hydraulic metal and plastic molding, pressing, extrusion devices, and all traditional mechanical metalworking and plastic shaping and molding machines and tools. 
It could also theoretically mean that genuine mass-production of rocketboosters, spacecraft and components can be developed with minimal manpower, and with parts and hardware that can be defect free
(requiring no quality inspection by humans).

Not really.  3D only makes parts, there still is assemble and test.  And this goes for anything. 

Also, it doesn't guarantee defect free (no such thing) and still will require some quality inspection by humans.

You don't understand what it takes to make a launch vehicle or spacecraft if you think this will make them more mass producible.
Making brackets, fittings, boxes, parts for valves is not the issue.

[strangequark]

  This could ALSO mean the beginning of the end for metal lathes, reamers, hydraulic metal and plastic molding, pressing, extrusion devices, and all traditional mechanical metalworking and plastic shaping and molding machines and tools. 
It could also theoretically mean that genuine mass-production of rocketboosters, spacecraft and components can be developed with minimal manpower, and with parts and hardware that can be defect free
(requiring no quality inspection by humans).

Not really.  3D only makes parts, there still is assemble and test.  And this goes for anything. 
 Most NC machines are automated.  There is very littl ehuman involvement.

This wont change the space launch paradigm. 

Also, it doesn't guarantee defect free (no such thing) and still will require some quality inspection by humans.

Don't forget the part finishing too.

Yeah, overall it's a nice technology, and it has a lot of interesting applications, but it won't do away with traditional machining anytime soon. There is also a size limitation. Granted, that will probably increase over time, but you're never going to see primary structures printed. Can you imagine what an ET printer would look like ?

Good for small parts with complex geometries that benefit from being one piece. Turbine blades are a good example. Injectors and regen combustion chambers could probably benefit too.

I think the more exciting application, vis-a-vis spaceflight, is for eventual off-world use. Being able to print replacement parts for your machines may be pretty helpful for something like a moon or Mars base.

[Robotbeat]

In my post above, a regen rocket chamber is made using so-called "3D printing" technology. But it doesn't entirely replace all machining operations. In the image below, the blue and orange parts all represent areas where the regen rocket chamber needed to be threaded, since good threading requires much greater precision than 3d printing generally produces.

It's not impossible for 3d printing to create threads, but they're generally inferior to threads made if you use a tap and die set. Which is totally fine. It doesn't take that long to cut the threads.

Really, it's a quite interesting technology that we're just starting to see come into more practical use, but it won't replace traditional machine tools like lathes, etc. It's a good addition to them, not a complete replacement.

[kevin-rf]

Didn't we recently have another thread on 3D printing? http://forum.nasaspaceflight.com/index.php?topic=25091.0

Good for small parts with complex geometries that benefit from being one piece. Turbine blades are a good example. Injectors and regen combustion chambers could probably benefit too.

Um, aren't turbine blades mono crystalline?

Though, I will say the New Scientist article was interesting, and I was wondering how long before it would appear on NSF.

Btw. Hobbyist plastic 3D printers can now be had for under $1000 US. Google RepRap (http://reprap.org/wiki/Main_Page)

[strangequark]

Um, aren't turbine blades mono crystalline?

For the high pressure turbine in many jet engines, yes. However, there are other applications. I'm not intimately familiar with design requirements for LPTs, but I could see DMLS parts being acceptable, and easing manufacturing.

I do know for a fact that GE Aviation is doing R&D on where they could use DMLS.

[MP99]

There is also a size limitation. Granted, that will probably increase over time, but you're never going to see primary structures printed. Can you imagine what an ET printer would look like ?

Exactly my first reaction. But then I began to wonder about laying a tank down using a method similar to coil pots in clay (see attachment).

Start at the bottom of the tank, perhaps with a pre-formed sump. Place on a turntable, and use something akin to a printhead to lay a spiral of material, one very thin layer at a time. The head only needs to be as wide as the thickest element that needs to be produced. As the turntable rotates, the head lays a continuous layer, sintering as it goes. The head only needs to move radially from the axis (like the head on a hard disk drive) and upwards as more of the tank forms.

I presume a tank that is designed to support it's own + upper stage + payload weight while upressurised (ie not a balloon) could support itself through the assembly process, since it will be vertical and rotationally symmetric at all times. Obviously, it will need external support to hold it upright and ensure it maintains the correct shape during construction.

The "print head" would need some way to support elements that jut out from the basic structure between laying the powder and sintering, eg stringers etc. Perhaps the bottom dome could be supported from underneath by the "traditional" powder method while being formed. I believe that masonry domes can be constructed by adding one layer of bricks at a time which then become self supporting (circular structure in compression), so the same might apply for the top dome - only need to support a circular section where the rim is being formed. Alternatively, print the top dome upside down, then weld or sinter to the top of the otherwise complete tank.

The "print head" could also include instrumentation (ultrasound, x-ray, etc) to continuously check the structure as it is laid down.

Re the thrust structure where it joins to the domes / tank barrel, I wondered whether the cylindrical section could be formed first, then moved away whilst the dome is formed. Once the diameter of the dome matches the inner diameter of the thrust structure, place it concentric with the dome then add the layers to join the two structures together. This also becomes the structure that supports the tank walls as they are constructed above. This would probably require that the top dome / thrust structure be formed upside down with a short barrel section, then attached to the top of an otherwise complete tank.

cheers, Martin

[KelvinZero]

IMO the really staggering potential of 3d printing for space is the step towards self sufficiency using local resources. The amount of infrastructure we would need to put on the moon to build a simple washer or screw the way we do on earth would be staggering.

It is ironic that when we talk about space-age technology and materials we are usually talking about cutting edge technology that would be very difficult to manufacture anywhere but earth.

IMO learning to live self  sustainably on other worlds will be all about learning to use insitu resources to manufacture components  that may be quite shoddy, using quite labour intensive approaches, compared to what we can do on earth. They merely have to be good enough.

Building rockets is almost the last thing I would worry about. Once you are on the moon with access to heat, cold and volatiles there is an awful lot to explore and master before turning lunar resources towards rockets. Can we keep our lifesupport in repair? can we make a brick? An electric engine would be much more interesting to me than a rocket which is after all just a way to throw away lots of volatiles to go somewhere else when you have everything you need to live and grow right there.

[Tcommon]

...Now, according to an article in the science magazine, NewScientist (issue: July/30/2011), a recent type of industrial production called 3D printing (still in the experimental stage) promises to change everything according to Paul Marks (the journalist who wrote the story).

Bla. 3D printing this has been around for decades, including sintered metal processes.

A while back P&W had a blog release on J2X development that included a printed metal tube. They hailed it as a sterling example of the J2X's stunning application of new technology. The particular process they used has been around for over twenty years. They hired a jobshop to push a button and didn't even print the full size part - had to do some welding afterwards. Some amazing new technology there.

[JohnFornaro]

Decades?  I'm ordering my 3D printer next week.

Didja see the August issue of NTB, p.31?  The Roland MDX-540, starting at eight grand. 

[Lee Jay]

IMO the really staggering potential of 3d printing for space is the step towards self sufficiency using local resources. The amount of infrastructure we would need to put on the moon to build a simple washer or screw the way we do on earth would be staggering.

There's not much that's simple about a screw.  An ordinary grade-5 screw you might buy at a hardware store has rolled threads, a broached head, is hardened to a pretty high level, and is likely treated in one way or another to be corrosion resistant.  Making one using 3D printing will require an extreme level of precision, and several operations afterwards.

[JohnFornaro]

There is also a size limitation. Granted, that will probably increase over time, but you're never going to see primary structures printed. Can you imagine what an ET printer would look like :o?

Exactly my first reaction. But then I began to wonder about laying a tank down using a method similar to coil pots in clay ...

Are you in my mind?  Because I had a very similar idea: squirting out the coils of a regen nozzle along the proper profile.  Hadn't thought about the ET, tho.

The other idea I had was to grow the hull of a spacecraft, embedding sensors and ethernet, say, along the way.  I took embryonic growth as my model for this.  Then the drugs wore off.

It is the idea of printing small titanium parts, say, on a lunar base that would be quite useful.  But shoot, the technology would be the enabler of a whole nother industry here on Earth.  The inside visor clips on my '92 Volvo broke.  They're simply not available, that I can tell.  Wouldn't it be nice to temporarily glue the pieces together, scan them, tweak the part on screen, and have your printer make it?  I could make a new, improved map pocket for the passenger door.  The central console, including cd storage and a drink thingy which wouldn't spill on the cd's.  Fabricate a new seat belt retainer for the third seat.  Maybe other people have better ideas of stuff to make, but hey.  I'm prosaic that way.

[KelvinZero]

There's not much that's simple about a screw.  An ordinary grade-5 screw you might buy at a hardware store has rolled threads, a broached head, is hardened to a pretty high level, and is likely treated in one way or another to be corrosion resistant.  Making one using 3D printing will require an extreme level of precision, and several operations afterwards.

Um.. I don't doubt it

Maybe it would always be worth having a specialized machine just for screws at that. I guess a complex integrated circuit or solar cell might be easier than a screw.

[Robotbeat]

...An electric engine would be much more interesting to me than a rocket which is after all just a way to throw away lots of volatiles to go somewhere else when you have everything you need to live and grow right there.

Here ya go:
http://www.thingiverse.com/thing:11164
(Using 3d-printed parts and PCB, some magnets, and some steel washers)

[Robotbeat]

You wouldn't want to use a 3d printer to print things like screws or threaded rods directly, though those are definitely important parts of a 3d printer.

HOWEVER, you can print off a mini lathe with a 3d printer (plus you need threaded rods and bolts for some parts), with which it is possible to cut soft metals like copper or brass (with which you could build all the parts for another lathe, plus the 3d printed parts). With investment casting using 3d-printed molds (or making molds of the 3d-printed parts), you could make a much stronger lathe capable of cutting steel.


A 3D printer doesn't negate the need for a metal workshop, but it can help you make one. But a 3d printer doesn't negate the necessity for assembly and finishing.

I agree that tools like this are useful for helping to enable self-sufficiency.

[Moe Grills]

  This could ALSO mean the beginning of the end for metal lathes, reamers, hydraulic metal and plastic molding, pressing, extrusion devices, and all traditional mechanical metalworking and plastic shaping and molding machines and tools. 
It could also theoretically mean that genuine mass-production of rocketboosters, spacecraft and components can be developed with minimal manpower, and with parts and hardware that can be defect free
(requiring no quality inspection by humans).

Not really.  3D only makes parts, there still is assemble and test.  And this goes for anything. 

Also, it doesn't guarantee defect free (no such thing) and still will require some quality inspection by humans.

Though I didn't quite explain it well, yet I do agree that there will always be the need for a human (or humans) in the loop, but my prediction was based on the assumption of increasing
automation and correspondingly increasing reliance on software (assuming
the CAD-CAM software fits this acronym, QIQO.) in the decades to come.   

[baldusi]

  This could ALSO mean the beginning of the end for metal lathes, reamers, hydraulic metal and plastic molding, pressing, extrusion devices, and all traditional mechanical metalworking and plastic shaping and molding machines and tools. 
It could also theoretically mean that genuine mass-production of rocketboosters, spacecraft and components can be developed with minimal manpower, and with parts and hardware that can be defect free
(requiring no quality inspection by humans).

Not really.  3D only makes parts, there still is assemble and test.  And this goes for anything. 

Also, it doesn't guarantee defect free (no such thing) and still will require some quality inspection by humans.

You don't understand what it takes to make a launch vehicle or spacecraft if you think this will make them more mass producible.
Making brackets, fittings, boxes, parts for valves is not the issue.

All very true points. But part of what's amazing of direct printing, is that you can make parts that are impossible to machine, and thus can actually save assembly steps, unions and seals. Personally the current precision, finish and thermal treatment isn't even close to what's usually needed. But for things like manifolds, turbo cases, regen chambers and injectors, the technology is extremely promising. There's still the point that even if you invented a magical machine that can make perfect and tested rocket engines, you'd still have to do a lot of assembly, integration and testing of the whole stack. But if there's a machine that can do the whole chamber+injector, and a single manifold+turbo body, it would slash manufacturing costs.

[Downix]

Decades?  I'm ordering my 3D printer next week.

Didja see the August issue of NTB, p.31?  The Roland MDX-540, starting at eight grand. 

I'm just building myself a reprap, so I can experiment with various technologies.