steps/mm dilemma: is more than 0.1mm/microstep necessary?

Quote
the_digital_dentist
The final pulleys are mounted on the machine and working fine, but the application isn't nearly as backlash-critical as yours.

What about the larde as the printer footprint disc with friction gearing from a stepper motor? Am i right in my assumption that large gearing ratio will make it possible to engage the gear and pinion with relavively little friction (compared to what a 5:1 gearing ratio would offer)? This will be as backlash-free as it gets.

On the other hand would such a big gearing ratio make backlash insignificant if I would use a herringbone gear profile like in geared extruders?

I would like to stick with spectra for it's backlash-free properties but for gearing down the stepper motor on Z it doesn't really make sense. The most appealing to me is the friction drive. They are turning large telescopes with it.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

What is "friction gearing"? If you mean just wheels pressed against wheels, it may eliminate backlash but it may slip. I can see where meshing teeth could be a problem if you're trying to take pictures through a telescope.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
the_digital_dentist
What is "friction gearing"? If you mean just wheels pressed against wheels, it may eliminate backlash but it may slip. I can see where meshing teeth could be a problem if you're trying to take pictures through a telescope.

Sorry about my previous posts' messup. No idea how it did end up like that.

And yes, I was talking about a large disc (we can basically have a lazy susan-like disc that is as big as the printer's footprint) and a little friction disc on the motor. But you are right, I am trading a fraction of a tooth as backlash to maybe several stepper turns as slippage and I wouldn't know it.

But, if properly adjusted, it shouldn't happen. The chance of wheels slipping is as big as the chance of motor skipping steps and that is again something the controller is unaware of. I was thinking that a large gearing ratio would somehow make sure that slippage will not happen due to the relatively small needed torque to turn the large wheel.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

Back in the old days, cheap record players used friction drive like that- the platter was made of steel and the motor had a steel wheel with a rubber coating. Over time the rubber always hardened and eventually cracked and naturally the record player started sounding pretty bad. Or it would develop a flat spot from sitting for a long time with the same result. Later, that sort of thing was replaced by belt drive- metal pulley on the motor, metal platter with a groove cut for the belt. The belt was usually just a large o-ring, and it had to stretch a little to fit over the platter and motor pulley. Besides performing better, the belt was easily user replaceable. In good turntables, they used to make the platter as massive as possible so it would have large rotational inertia.

The large wheel would somehow be coupled to screw(s) to lift the bed?

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Much more experimental than that ... I can't even explain but when i have a prototype i'll post it somewhere on this forum. It is some of the Z solution in RichRap's Slid3R, some friction-drive reducer and some reciprocating screw that will prevent spectra walking. If I can manage to: 1) make the friction wheel work without slippage and 2) find a way to avoid lateral play between nuts/screw, I'll have a very snug spectra-based Z lift, which I like because on the SCARA stage of Lander there is no backlash at all.

1) the friction-drive might be backed up by a herringbone gear setup so that both are engaged at all times and if there is any slippage, it shouldn't be more than the remaining of the distance between the engaging tooth and the next one. I am also betting on the "huge" factor, a large gearing ratio will lower the slippage chances a lot. Before you ask, it was a goal to "avoid belts until we can print them" ... I am still trying to live by the original goals

2) There are very low chances I'll be able to prevent lateral play in a normal bolt/nut assembly so I am still pondering on the reciprocating/sliding solution.

Edited 1 time(s). Last edit at 11/11/2015 08:52AM by realthor.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

Ok, not that my current design would n't be complicated enough but I was never happy with the solution above. It was going to take the complexity to new heights, so I am happy to have reverted to the same approach I have with the arms of the bot but applied for Z stage. The solution was there all along but I just couldn't see how to use it. Now I have a 15:1 reduction before microstepping, so 150steps/mm with no microstepping... I might go 4x or 8x. So no backlash and enough torque for the heavy/large printbeds that scara are capable of using.

Regarding the question in the thread, I still don't have a satisfying answer as to why we need more steps/mm than our nozzles can handle in layer height. One possibility is that all the accumulated errors in the mechanics will make it easier to work with if we can do small increments in Z height. This might be enough to shoot for 2000steps/mm but there might be more.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff