Experimental Off-Carriage X-Axis

I recently had an idea about moving the X axis motor off the X carriage. I suspect there is little benefit of doing this, but I felt I had to at least knock out a quick prototype to see it in action.

The video below actually details it all quite well, but i’ll also add a few pictures and a description below.

So, the idea is to move the X motor off the carriage. One immediate benefit is that the size of the entire X axis is no longer constrained by the size of the motor, perhaps leading to more compact designs?  It would be interesting to try out a vertical X axis design with this setup.

Anyway, the X motion is transferred to the belt via a square extrusion rod which drives a cog which sits directly on it.  This rod slides through the mount and cog during Z motion.  The belt in this design is within the two X smooth rods, but other layouts could be adapted to move the belt to the outside (as with  the original Huxley I believe).  In this design the hot-end would either have to fit through the belts, or hang over the edge of the rods.  Again, here the vertical layout would be interesting.

One comment

  • September 27, 2011 - 11:29 pm | Permalink

    Very interesting! It might work better, though, on a machine with a Darwin-style X/Y axis, so that it could take weight off of a fast-moving axis instead of the slow-moving Z-axis. I’m designing a bot of this type and I’ll add this to the list of mechanisms I’ll consider to get the X motor off the Y axis.

    OTOH, if you also used a bowden extruder to take the extruder motor weight off of the Z-axis, this should make that axis light enough to use a belt drive instead of leadscrews and make it just as fast as the other axes. Not sure why you’d need that since the Z axis only needs to move in small, infrequent steps during a build, but it would be a cool capability.

  • Comments are closed.