[nickpinkston]

Good to see the Economist covering 3D printing - it's still amazing to me how much most people, even close to manufacturing, aren't familiar with it at all.

FWIW though - the 90% they're referring to is the delta between machining solid parts and printing them. It should be added that this is a rare use-case for 3D printing.

Usually it competes more directly with injection molding technologies that often have better material usage because there is no need for support material - which goes to waste holding up the hollow areas for the types of material-saving lattice structures they're talking about.

Sometimes you can get great savings, but 3D printing isn't a production materials panacea. Let's not even get into how to recycle composite materials (very energy intensive) and photopolymers (you can't).

[Kliment]

Well, that's not entirely true. The lattice structures are internal to the object, and substitute for solid infill. Commercial FDM 3d printers use a crosshatch instead of solid fill, and the reprap toolchain has that as well as a number of cool ones like hexagonal infill. So you have lattices instead of solid walls in the microstructure. With powder printers, such as the metal printers described in the article, you can form hollow areas of any shape since the raw material itself acts as support. To get the material out they need to be open on one side, but that is hardly an issue with a design that airy. No additional material is needed. The only case where you need support is for FDM based printing with sharp overhangs. Even this can be mitigated with clever design, such as making overhanging parts with bridges (connecting walls with a stretched layer of material that spans a gap) or teardrops (closing gaps using low overhang structures, like the buttresses on gothic buildings). The main advantage of 3d printing is that it can produce structures internal to the object that cannot be made by any subtractive method, and that it makes production costs linear and design-independent. This is very cool, since it literally costs the same to make two identical or two different objects (with the same build time) and it separates cost from volume. The costs per unit are still high compared to injection or casting, but the tooling costs are essentially zero. And the costs per unit are dropping rapidly, particularly given that the FDM printers themselves are now partially self-replicating

It's a very exciting time to be living in. I have three printers. The first printed the second. The second printed the third. I've made a number of things which turned from idea to design to prototype within hours to days. Then I tweak, hit print again and I've got production. No tooling, no waiting.

[nickpinkston]

Actually, FDM, Objet, ProJet, SLA, EnvisionTec, SolidScape, an many more use support material that is consumed during the build / post-processing.

Many lattice structures can still be made using injection molding equipment (far cheaper) or starting with a raw material with such a structure: think of aluminum honeycomb. 

I agree 3DP is exciting, but we need to get real about what we're trying to achieve - traditional manufacturing engineers are pretty clever too. Not every STL should be run through a nozzle... 

FDM is actually one of the worst methods for 3DP manufacturing because it scales linearly - MakerBot et al use this tech because it's the easiest process to build and makes durable parts from common raw materials.

[Kliment]

Indeed, but with clever design you can avoid support material during builds and still get adequate results.

I fully agree that you can get lattice structures in a number of processes, but printing is the only reasonable way to get internal, closed lattice structures. And I believe the article was meant to create awareness of a reasonably unknown process, rather than claim that everything else should step aside. Milling, routing, turning and molding aren't going anywhere. But that doesn't make printing any less impressive in the areas it's good at.

FDM scales linearly, which is a good thing and a bad thing. It means you have no cost savings when you make large quantities of the same object, but it also means you have no additional cost to make a different object every time. That's what I think is most exciting about it.

[nickpinkston]

Yea, I only mention these things because there are so many "OMG 3DP is good at everything!" fanboys out there. The open source FDM machines are very exciting. I still wonder what will happen to Stratasys when MakerBot et al get soluble support and finer nozzle size. Should be interesting.

[michaeldhopkins]

Did you design your printers or buy them? You should write this up and submit to HN.

[Kliment]

Bought the first from makerbot, was a huge pain and disappointment, but it got me as far as printing the parts for the second one, a Reprap Mendel. That one in turn printed me (and a number of other people) the more streamlined Prusa Mendel. I'm a contributor to the Prusa variant (I wrote the build documentation and the makefiles, and contributed to some of the parts), but did not design it. So yes, as others said below, I have two repraps (and one non-reprap), and I did build them myself.

[michaeldhopkins]

Thanks. I was thinking about getting a Makerbot. I'll look into these others.

[Kliment]

Strongly advise against the Makerbot products. They cause the highest amount of headaches and are riding on mindshare and marketing rather than quality. Have a look at http://repraplogphase.blogspot.com/2011/01/death-of-sub-1000... for how it came to be this way.

[burgerbrain]

It sounds like he probably has 3 RepRaps.

[chronomex]

Kliment is active in the #reprap channel on FreeNode, so it's safe to say that he built his own printers.

[StavrosK]

How much work did it take to create a printer when you already had one, and how many custom parts?

[Kliment]

About 20 hours of print time, 4-5 hours of assembly time, and around 400EUR of nonprinted parts (motors, electronics, fasteners, hotend parts)

[StavrosK]

Oh, very nice! That's quite handy.

[podperson]

I was a bit curious about the 90% figure, but a good deal depends on what point in the adoption process we're talking about, e.g.

Use 3D printing instead of milling or whatever for prototype.

Use 3D printing instead of whatever for mass production.

Use 3D printing at point of use.

Simple example: I bought a porcelain teapot a few weeks ago. A few days later we broke the lid. So I bought a new teapot. (The old teapot with the broken lid is sitting on a window sill above the kitchen sink, waiting for glue. A tube of glue costs a substantial fraction of the price of the teapot.)

3D printing could eventually lead to a tectonic shift away from planned obsolescence and disposable culture (which, let's face it, is purely an artifact of mass production).

[nickpinkston]

Yea, I'm very much with you. Our goal (at my startup CloudFab) is that by making things for ourselves - or at least closer to what we want - that it'll mean less marketing driven consumer culture and people will want to keep their self-designed goods.

FYI - we can print in ceramics now - so that lid is replaceable if you've got a file ;-)

[bobds]

As far as recycling goes, it's very exciting that you can recycle plastic bottles into plastic granules/filament and use them for printing.

I think we are not that far away from a self contained fabricator that turns garbage into everyday plastic parts you need.

[chadgeidel]

It's probably going to take a little more than that, but I've always held that landfills are probably going to be very valuable property in 50 years or so. I assume there are going to be "dis-assemblers" to consume and separate "junk" for repurposing in manufacturing.

It's a little dream I have. :-)

[patrickk]

Your vision is like a scene out of Wall-E!

[tintin]

Well I'm not sure you could say the same for this example: http://www.youtube.com/watch?v=zwHgszH0aqI (printing a working flute, multiple plastics at the same time).