the Will Langford links have been hacked - don't go ther

Thanks. I updated the links to point to wayback machine links.

It seems I haven't had time for this project.

micron level tolerances are the norm with wire EDM.  i think you're going to be disappointed if that is what you're going for.  Also regardless of how you get the arms positioned.  You are still talking about a straight line (the wire) between two points.   I would think you would be better off (unless the point is to play with robotic arms ;)) with an x-y on top and bottom to position the wires.  holding the workpiece you could have a "lathe" that can index or rotate.

This is ridiculous, but I'm going to work on the double 6DOF articulated robot arm wire cutting system design because I think I can eventually figure out a way to achieve the accuracy and precision I'd like from the system, and because I want experience writing code for these kinds of robot arms. I bought the least expensive 6 DOF robot arm I could find that looked like it will survive more than a month of testing in order to experiment with motion profiles.

I'll use this post for links that I think are helpful or relevant for reference:

A Parallel SCARA robot design:  http://forums.reprap.org/read.php?185,490216,page=2

SPECIFICATIONS

------------------------------------------------------------------------------------

I'll propose project requirements, and I'd like to know if anyone would like to recommend different specifications for the design.

I need to do some homework on definitions for repeatability, accuracy, precision, and more; but I want the system to be able to achieve a machining tolerance of +/- 0.0001" (0.00254 mm) in the X, Y, and Z directions. 

As for the build volume, I personally don't need to make components larger than 5" x 5" x 5" (127 mm/axis), so I'll start with this.

Using two articulated robot arms (think Baxter robot), would provide much more flexibility in the possible wire cutting positions and would be extremely fun to make.

I'll stick with the parallel SCARA design for now because I think it will be difficult to achieve high accuracy at a low price because of tolerance stack up from the backlash in each joint and because of the cost of additional motors, controllers, and assembly components.