Re: EMC and 4th axis

Dan Mauch wrote:

> I looked all over and I can't find any info on adding a 4th and possibly
> fifth axis.
> But the software says it supports up to 6 stepper motors.

Here's the scoop on 4 or more axes:

1. I recently reworked the EMC motion controller so that it more cleanly
separates Cartesian coordinate from axes. This was to support a small
6-cable Stewart Platform "robot crane" driven by steppers. If you
haven't seen the NIST Robocrane (tm), it's like a hexapod machine only
with cables. The cables are fastened in pairs 1-2, 3-4, 5-6 at the top
of a triangular frame, then attached to stepper motors in pairs 2-3,
4-5, 6-1 on a smaller triangular platform that moves around.

I put the EMC on this and cut a circle-diamond-square test program in
stryrofoam using a Dremel tool mounted on the moving platform. The
kinematics are very strange, but the EMC supports cable jogging/homing
and coordinated-cable linear moves, circular moves, etc.

2. I wrote a slightly different version of the xemc GUI, that displays
all 6 joint (cable) lengths, or XYZ roll-pitch-yaw. There's a menu item
for displaying these as Joint or World frames, respectively.

3. The EMC still only does Cartesian motion planning, so the
roll-pitch-yaw numbers don't change. I'm working on this with a student
from Stanford now. This should be done by the end of September.

4. The RS-274 G code interpreter only handles XYZ, but we have other
versions that we've written to handle XYZB and XYZAC formats. These need
to be integrated into a single one (or at most two) that handles XYZABC
(or XYZIJK for hexapods). Still working on this also. This should be
done in a couple of months, as soon as Tom Kramer, G code man, gets some
time.

In summary, you'll have to wait until we get full 6-degree-of-freedom
motion planning, and an interpreter that can feed it.

I'm still working on feedback for steppers, a new motion board, bug
fixes, ...

--Fred