Prusa Z axis coupling reliability

Hello All.

Is the prusa z axis coupling with the motors reliable? seeing the motors connected from the top seems weird. I currently have the huxley and the motor is connected from the bottom. Too much force on the motor... no?

Gary had an issue with that: [garyhodgson.com]

---

[omemon.wordpress.com]

I should clarify, because reading my post again now it isn't very clear, but the breaking of the coupling was completely my fault - I was putting too much pressure on the entire x carriage as I was trying to push through some blocked filament by hand (I know, I know, i've learned a lesson in patience now).

So far the Z axis has been holding out well for me (touch wood!) apart from a slight wobble in it which I haven't gotten around to correcting.

I'd be interested to hear if the benefits of bottom z axis motors is worth considering for future Mendel mutations?

Edited 1 time(s). Last edit at 04/27/2011 11:18AM by garyhodgson.

The coupling in the photo looks like it has sparse infill. I print those 100% solid because they are quite fragile when sparse.

---

[www.hydraraptor.blogspot.com]

Thanks guys, this helps me prepare which reprap i want to build next.

Anyone know about the wobbling issue with the couplings?

---

[omemon.wordpress.com]

The Prusa couplings are a very weak design. They don't flex, and getting the motor shaft and threaded rod to line up just right takes luck and repetition. PrusaJr recently created a two part coupling that I like much better. I imagine it will be making its way into the master branch soon enough.

I tried my hand at designing a flex coupling, but it does not really flex at this point, and is really tough to print. Everything needs to be thinned out to get flex in PLA. If anyone wants to have a look, check it out on Thingiverse

[www.thingiverse.com]

Nice design, test it out and see. I saw some people on the comments said its too flexible

That leads me to this idea: A universal joint: [en.wikipedia.org]

or others listed here: [en.wikipedia.org]

Would that be mechanically correct? Any stress issues there?

Edited 1 time(s). Last edit at 04/27/2011 02:30PM by owism8.

---

[omemon.wordpress.com]

The problem with anything flexible is it will be flexible in the Z direction where you want it absolutely free of any play, so you would need a bearing after the coupling. The original Mendel design is much better in this respect.

The key thing is to get straight rods so it doesn't need to flex.

---

[www.hydraraptor.blogspot.com]

I have had many bouts with this issue.

I cut teeth 2 ways on my motor shaft, then found that PLA couplings tightened down seemed to grip on better than ABS.

The comment about being too flexible was pertaining to another design on thingiverse (not mine). That design resembles a spring, which is not desireable for suspending the entire X axis.

My design is supposed to flex laterally in X/Y, but Z is absolutely rigid. This allows the threaded rod to live peacefully, even if it is misaligned with the motor shaft. The problem with the current incarnation is that the flex points between ribs are too thick, and don't allow any flex at all. I'll probably take another stab at fixing that this weekend. It's been languishing while I build up a second machine.

I actually have this same design working on my machine, but I used Acetal Acetate which I formed using a small hobby table saw. Maybe I should post that up as a derivative...

So what pros are there to this "2 motor on top" design? Why not just switch back to old mendel days of one motor on the bottom with two gears being turned by one belt.

This pro/con should be on the wiki under Prusa if anyone has time for that.

I appreciate the discussion fellow reprappers.

---

[omemon.wordpress.com]

The pro is that it's extremely simple, and possibly cheaper, depending on the belt.

I think there is a good half way solution: Anchor the threaded rods from top to bottom to be stable, and not have gravity work against them etc. + drive them with a printed gear from the motors, in other words, off set the motors to printed gear drive more stable threaded rods, this removes the pulley issue, and the z rod wobble issues, in theory of course.

You could try my coupling: It uses 3 M3 screws as a 3 point vice to grip an M8 nut locked onto the top of the z rod. I drilled small dimples with a 4mm drill bit in 3 sides of the nut for the M3 screws to sit in. Once you clamp the nut, you can spin the z rod by hand to see if it is alighed, and adjust the 3 screws if necessary to align it:

[github.com]