Open-Source 3D Printer Lets Users Make Anything

An anonymous reader writes "Picture a 3D inkjet printer that deposits droplets of plastic, layer by layer, gradually building up an object of any shape. Fabbers have been around for two decades, but they've always been the pricey playthings of high-tech labs — and could only use a single material. A Fab at Home kit costs around $2400 and allows users to print anything from Hors d'Oeuvres to flashlights."

More Discussion

[eldavojohn]

You probably remember discussing this almost a year ago [slashdot.org]. Enjoy more on this at that coverage of the same story.

Re:Any shape?

[LiquidCoooled]

You solve a problem like this by laying down sand or another substance to act as the free space and support the structure.
After building you remove the sand and your 3d model emerges.

RepRap Is Even Cheaper

[Dean Edmonds]

A RepRap [reprap.org] machine costs less than $500 in parts, though it does require a lot more assembly work.

Re:Any shape?

[FishWithAHammer]

Yes, but what about hollow objects, like an egg?

Squeeze bulb? [fabathome.org]

Re:Manufacturing is a solved problem

[PieSquared]

No it isn't. We still have one really major step to take (that we can see from here/now). Molecular level construction. I don't mean nano-tolerance specs for things this could print, but by building things at the molecular level you finally get the ability to do self-replication. Right now the problem is that you need a scale - if you have a stick you use to measure things by, you add to the error of *every* measurement with each generation... which prevents self-replication. If you use a molecule (or some universal constant) as your stick, though, you lose this problem... stack a certain number of molecules into a stick of a useful size, or use the speed of light (in some medium) to measure distance for your "unit length" as part of the replication process and you'll have the same error in every generation. We already do this for manual manufacture... just, because we don't try to make self-replicating fabricators, we only have to measure (using the standard of the speed of light in a vacuum) once every few years to replace the "standard" used in manufacturing rulers.

Molecular level construction could also be useful for, obviously, building really small things. Or for building really big things semi-automatically.

Once you can spec the atomic placement in manufacture.... *then* there will be no need for brains in manufacturing. That we can understand today. Who knows, maybe there is something useful beyond that level that we just don't understand yet. But for now this is the one major step left in the ability to manufacture things.

Re:I'm not convinced...

[monopole]

Erm, Rep Rap [reprap.org]
I know, it won't fab everything but the few remaining bits are easy to get.

Re:material

[Trogre]

The RepRap project, while currently using Polycaprolactone, aims to eventually move over to polylactic acid from corn starch or sugar cane.

Re:Any shape?

[jank1887]

commercial software with 'support material' will look at overhanging structures. If the vertical angle is larger than a set value (maybe 45degrees) it will build a support structure under it as it builds. If the angle is less than that (as in the aforementioned squeeze bulb) it will be considered a 'self supporting angle. Enough of the upper layer bead will be on top of the lower layer bead to prevent it from toppling. This usually takes a bit of intuition, however, because simple rules like this will let you build the leaning tower of Pisa at too steep an angle for it not to fall over. (shifting the center of mass outside the footprint)

Re:Any shape?

[jank1887]

Anything with a laser is much more expensive (with fewer material options) than what's being discussed here. You are referring to a Stereolithographic [wikipedia.org] process, primarily commercialized by 3D Systems, Inc. [3dsystems.com]. This group uses more of a heated extrusion, similar to the Fused Deposition Molding [wikipedia.org] process used by Stratasys, Inc. [stratasys.com] Even the liquid resins, though, have limits to degree of overhang permitted before the cured material will sag or fall in.

Re:Can I make a 3D fake pussy?

[Anonymous Coward]

You could probably print a Bush mask. He's a bit of a cunt.

Re:material

[Anonymous Coward]

something like this:

http://www.mcp-group.com/rpt/rpttslm.html [mcp-group.com]

Re:Any shape?

[ShieldW0lf]

Early 3d fabs for medical use used this method. They would, for example, recreate a broken skull from x-rays and let the doctor practice putting his hands in it before surgery.

I've been thinking of sinking the money into getting parts for a Rep-Rap [reprap.org]. These look nice though.

Not ready for prime time

[seven of five]

I was pretty excited by this as earlier reported, but looking into it for a while, realized that you can't do precision fabbing with one of these el cheapo machines, not yet. The blobs/droplets are too big and the stepper motors spec'd at this price don't have the accuracy either. This will improve with time but 'not yet'.

Re:I'm not convinced...

[CarpetShark]

Actually, they're not. Yes, I tried. I used the new shop which is supposed to provide an even easier way to get components. Package arrived, and I realised later there were LOTS of parts missing, although I ordered everything I obviously needed on first read. Seeing the ACTUAL list of parts was a shock. It goes on forever. I gave up. RepRap is a LONG way from being usable by the average joe, and the RepRap team will probably tell you as much.