There is an obvious type of kinematics that would almost certainly be terrible for a 3D printer because of its asymmetry, but might be good for a laser cutter, and there's a lot of overlap in technology between the domains.

But I don't know what it's called, and it's sufficiently obvious that I expect it to have a real name.

Here's a simple schematic diagram.



The Y motor has normal cartesian motion. It's moving a gantry back and forth. The X motor is moving a head/effector, and it needs to compensate for any Y motion by composing the inverse of the Y motion.

• If only the X axis is moving, only the X motor turns.
• If only the Y axis is moving, the X and Y motors need to turn equally in opposite directions.
• To move on one 45⁰ diagonal, the X motor would stop and the Y motor would turn.
• To move on the other 45⁰ diagonal, the X motor would turn twice as fast and in the opposite direction as the Y motor.

It's not one of the kinematics supported by RepRapFirmware, Marlin, Smoothieware, or Klipper as far as I can tell.

If anyone knows an established name for this, I'd be grateful to learn.

It's commonly called MarkForged kinematics. It's supported by the standard builds of RepRapFirmware, along with every other kinematics for which the movement of every axis is a linear function of the motion of the motors.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Thank you! ☺

I just read the docs for M669 and that's wonderfully general.

Edited 1 time(s). Last edit at 08/29/2020 05:13PM by mcdanlj.

much better if you rotate the arrangement 90 degrees

Quote
MechaBits
much better if you rotate the arrangement 90 degrees

To make the middle axis shorter? Exactly what I thought. It cuts the deviation on the end point of the axis in half at least.

---

http://www.marinusdebeer.nl/

I think it makes sense when the axes are already very asymmetric, as is typical in a laser engraver/cutter. X is typically longer, possibly much longer, than Y, and rastering/engraving is typically done X-major. (Cutting is slow enough that intertia mostly vanishes as a concern relative to motor torque.)

In at least the case where I've been lightly considering it (for a laser cutter), the Y axis is actually driven at both ends, coupled by a torsion rod; that was just unnecessary detail for the schematic. I would think the same could make sense for a high aspect ratio 3D printer perhaps, if it was long enough belt length started to become a concern for CoreXY? Or for any other reason you didn't want to use CoreXY and still wanted to avoid stacking one of the X/Y axis motors?

But mostly it was intellectual curiosity, which many thanks to dc42 for satisfying. ☺