4 ms·

I'm not really sure what relation the article's headline has to its content - apart from the obvious problem that current 3D printing technology is nowhere near

by ZoFreX 14y ago

I'm not really sure what relation the article's headline has to its content - apart from the obvious problem that current 3D printing technology is nowhere near accurate enough for the tolerances involved in a jet engine, I don't think it will ever be possible to print one entirely, as the fan blades are each a single crystal of metal, and that's not a property you can achieve through any 3D printing method (or any incremental building method - they need to be created in one piece)

powertower 14y ago
I was thinking the same thing. They are most likely printing things that 1) don't move, and 2) aren't subjected to any type of stress or heat. I can't imagine that you can print something layer by layer with fused powder and expect to make anything as solid and tolerant as the normal (crystal) growth and/or casting/molding/etc processes make.
- guard-of-terra 14y ago
  In fact powder metallurgy is not new, it's an established and respected way to make various parts including ones for aerospace. http://en.wikipedia.org/wiki/Powder_metallurgy http://en.wikipedia.org/wiki/Powder_metallurgy
- stevenrace 14y ago
  NASA is set to launch in 2017 a rocket stage that was in part '3D printed' via a form of laser sintering. [1] Additionally, many aerospace parts (F22 air vents, for example) are produced in a similar fashion. There are a series of Google Tech Talks on the subject [2]. In short, complicated shapes can be produced with less waste (versus milling from a larger block of material), greater tolerances (no warpage from welding heat, curing of glue), less labor (no assembly jigs), and usually less mass (due to partial infilling of material cross section). Lastly, you'd be surprised at the number of molds that are now being made via '3d printing' (I'm growing to hate this term) for composite applications. Normally they are CNC'd from aluminum (or 'tooling gel') - whereas smaller run items can be laser sintered. My startup uses PLA molds for vacuum infused carbon composites which we print from a RepRap Mendelmax. (router enclosures, antenna mounts, UAV and motosports stuff, etc). [1] http://www.engadget.com/2012/11/09/nasa-building-space-launch-system-with-laser-melting/ http://www.engadget.com/2012/11/09/nasa-building-space-launc... [2] http://www.youtube.com/watch?v=s9dZQdVsBNA&feature=plcp http://www.youtube.com/watch?v=s9dZQdVsBNA&feature=plcp
hatcravat 14y ago
It's not so much that they need to be created in one piece [1] as that they need to be annealed after taking shape.
jamesmcn 14y ago
You can manipulate individual atoms with lasers, and 'cool' them down to nearly absolute zero. So I wonder how far we are from being able to grow a single crystal part via laser sintering?
- jacquesm 14y ago
  A very very long time. Laser sintering does not typically work on the per-atom level (those are called laser tweezers iirc), it works on relatively bulky material (powder or grains). The result is definitely not crystalline, in fact it is better compared with regular sintered materials.
nickpinkston 14y ago
So they're making net-near-shape parts in lieu of castings so they can: 1.) get different part properties (lattice structures), 2.) so they don't waste very expensive materials that are often blocks of unobtainium with 95% machined away... You're right - it's going to be very difficult to get high tolerance parts using any method of 3D Printing. Don't say never - but isn't not as close of the koolaid machine would have you believe.
jacquesm 14y ago
Fan blades are definitely not all 'single crystals of metal', though that is one method of manufacturing them. Usually they are simply cast and then reworked, which produces a blade of lesser quality than the single crystal ones but they are still quite usable (and a lot cheaper!).
- neurotech1 14y ago
  The newer fighter engines (F414, F119, F135) use a single piece "Blisk" fan and compressor, although I don't think its 'single crystal'. For prototyping, they can be 3D printed but are usually cast from molds and balanced for the actual engine.

by ZoFreX 14y ago

powertower 14y ago

I was thinking the same thing. They are most likely printing things that 1) don't move, and 2) aren't subjected to any type of stress or heat. I can't imagine that you can print something layer by layer with fused powder and expect to make anything as solid and tolerant as the normal (crystal) growth and/or casting/molding/etc processes make.

guard-of-terra 14y ago
In fact powder metallurgy is not new, it's an established and respected way to make various parts including ones for aerospace. http://en.wikipedia.org/wiki/Powder_metallurgy http://en.wikipedia.org/wiki/Powder_metallurgy
stevenrace 14y ago
NASA is set to launch in 2017 a rocket stage that was in part '3D printed' via a form of laser sintering. [1] Additionally, many aerospace parts (F22 air vents, for example) are produced in a similar fashion. There are a series of Google Tech Talks on the subject [2]. In short, complicated shapes can be produced with less waste (versus milling from a larger block of material), greater tolerances (no warpage from welding heat, curing of glue), less labor (no assembly jigs), and usually less mass (due to partial infilling of material cross section). Lastly, you'd be surprised at the number of molds that are now being made via '3d printing' (I'm growing to hate this term) for composite applications. Normally they are CNC'd from aluminum (or 'tooling gel') - whereas smaller run items can be laser sintered. My startup uses PLA molds for vacuum infused carbon composites which we print from a RepRap Mendelmax. (router enclosures, antenna mounts, UAV and motosports stuff, etc). [1] http://www.engadget.com/2012/11/09/nasa-building-space-launch-system-with-laser-melting/ http://www.engadget.com/2012/11/09/nasa-building-space-launc... [2] http://www.youtube.com/watch?v=s9dZQdVsBNA&feature=plcp http://www.youtube.com/watch?v=s9dZQdVsBNA&feature=plcp

hatcravat 14y ago

It's not so much that they need to be created in one piece [1] as that they need to be annealed after taking shape.

jamesmcn 14y ago

You can manipulate individual atoms with lasers, and 'cool' them down to nearly absolute zero. So I wonder how far we are from being able to grow a single crystal part via laser sintering?

jacquesm 14y ago
A very very long time. Laser sintering does not typically work on the per-atom level (those are called laser tweezers iirc), it works on relatively bulky material (powder or grains). The result is definitely not crystalline, in fact it is better compared with regular sintered materials.

nickpinkston 14y ago

So they're making net-near-shape parts in lieu of castings so they can: 1.) get different part properties (lattice structures), 2.) so they don't waste very expensive materials that are often blocks of unobtainium with 95% machined away... You're right - it's going to be very difficult to get high tolerance parts using any method of 3D Printing. Don't say never - but isn't not as close of the koolaid machine would have you believe.

jacquesm 14y ago

Fan blades are definitely not all 'single crystals of metal', though that is one method of manufacturing them. Usually they are simply cast and then reworked, which produces a blade of lesser quality than the single crystal ones but they are still quite usable (and a lot cheaper!).

neurotech1 14y ago
The newer fighter engines (F414, F119, F135) use a single piece "Blisk" fan and compressor, although I don't think its 'single crystal'. For prototyping, they can be 3D printed but are usually cast from molds and balanced for the actual engine.