I'm not sure if it's possible, the part that heats up the material can't be made from the material with the same temperature of melting as what it prints.
Ceramics or epoxy seem like to candidates. Something that has a curing step of some sort. Do you think you could make a kiln out of the same sort of material it is designed to fire?
Two or more printers that together can be self-hosting seems like a possibility to me. The only limitation would be size and price I think.
Didn't the old Makerbot print at 100 microns? The notes about the THE FREE UNIVERSAL CONSTRUCTION KIT http://fffff.at/free-universal-construction-kit/ referenced this as a limitation when trying to print LEGOs.
Both the Replicator and Thing-o-Matic could, but you had to get your printer tuned very well. The default was ~300 microns. If these come out of the box tuned that well, that's quite nice.
I don't think (though I'm not sure) the Cupcake was capable of that, at least with the stock Z axis.
Ultra-Bot on Kickstarter is about to offer an upgraded platform of 8x8x8" or 512 cubic inches. It has a very good resolution and can handle PLA and ABS thank's to the heated build platform. There is also a future option for a dual extruder.
For $1,099 you get an assembled Ultra-Bot with heated build table. Soon they will offer another, larger, Ultra-Bot for $1,249 with the larger 8x8x8" build platform.
No, it does not. The platform is made out of acrylic and apparently does not heat up. It looks like the Replicator 2X is designed to use both ABS and PLA but the standard Rep2 is PLA only.
I'm wondering how far of the ability to print out CPU. Certianly would appear that 3D printers will become viable for manufacturing more and more common day objects. I'm sure the ability to have more than one print head(ink/material if you like) and able to switch then things realy start to get very exciting.
> I'm wondering how far of the ability to print out CPU.
None. Current 3D printers can only print in a couple materials as of present (mostly plastics), none of which are semiconductors.
The resolution is nowhere near what you'd need for that, anyway. 100 micron resolution sounds pretty small, but it's still about a tenth the resolution of 70s-era semiconductors (10 µm), and about 4500 times the resolution of current tech (22 nm).
The time it would take me to design a CPU from scratch then I feel it is something that will be viable come that day I'm close to hitting print. But not today or any year soon then, but one day in my lifetime (cancer pending).
