RepRap, Lawrence Welk, and Bubbles, oh my!

Here's a blue sky problem I'm working on: a purely additive fabrication process in the spirit of the Open Source RepRap 3D printer movement, one that might come closer to fabricating all the components it needs to make a replica of itself with fewer "vitamin" inputs.

In the RepRap world "vitamins" refer to any components other than the feedstock material used to make 3D prints. For current RepRap machines these include precise steel rods, threaded rods, nuts & bolts, stepper motors, sensors, and all control and power electronics. The ultimate goal of a RepRap system is zero vitamin input, but all the current working systems are a long, long way away from this goal.

My idea involves using a 6-degree-of-freedom (6-DOF) "printer" head that would produce precisely placed and formed bubbles of some material that can be made to selectively adhere and harden in contact with previously deposited bubbles. If the bubbles are small enough and deposited in the right patterns it should be possible to print surfaces which can approach any arbitrary contour, perhaps even smooth enough to dispense with machined rods and screws.

The first problem, though, is to come up with an inexpensive material with which to make the bubbles. No doubt any number of exotic organic polymers can be found, along with numerous methods for their selectively adhering and hardening. The sorts of resins used for high-end stereolithography probably could be made to work with laser or UV curing, but those resins are quite expensive.

The goal is a cheap, versatile material, and ideally, for reasons I'll get to shortly, one that can withstand higher temperatures than are typical for organic polymers. My first candidate is a sodium silicate solution where the hardening agents are heat and carbon dioxide. The CO₂ would be used to blow the bubble to start the hardening from the inside surface, with a timed application of heated CO₂ "puffs" applied externally to finish the job. It might be possible to do a limited amount of "sculpting" of each bubble with the CO₂ puffs such that a smoother surface results.

So, now for the ideas I'm much less sure of. To make this sodium silicate foam useful, it would be nice to be able next to selectively plate (electroless or electrolytic) various metals on them. I'm thinking copper or copper alloys for making conductors or bearing surfaces, and nickel or iron or Ni/Fe alloys to make high-strength surfaces that can also have useful magnetic properties. If this can be accomplished then, in principle, you should be able to print any electromechanical component desired.

For instance, if among the feedstocks were nickel and chromium salts -- don't ask me which -- then you could deposit a nichrome heating element. The sodium silicate could withstand the heat (I'm hoping). It could also act as an electrical insulator, so use a different feedstock solution and you lay down the conductive tracks of a motor coil.

Here is where my ignorance will likely show the most, but here goes: I'm supposing a first electroless pass would be needed, followed by electrolytic passes to build up thickness. I'm further assuming that the electrolytic plating cell would be a small, continuously-fed "plating bead" held in place by surface tension between the plating head and the region to be plated -- no container, no large plating bath. The bead would be dragged along as plating progressed, perhaps encouraged along by puffs of air or CO₂.

Those are my basic ideas. Now the more knowledgable folks out there can point out all the flaws, gotchas, and implausibilities implicit in all this. I welcome the education, painful though it might be.

Also, to make sure I don't take the project -- or myself -- too deadly seriously, I'm calling it the "Lawrence Welk" RepRap. Or maybe just the "BubbleRap"...

-Bobby