Don’t get me wrong the zero already has a reasonable Z system but with a bit of tinkering I’m thinking a belted Z might be possible depending on the space underneath (not been under there in a while)
If the lead screws were replaced with rods and lmu’s it might be possible to shoe-horn a belted system in there.
Anyone ever attempted it?
I’ve never even thought about a belted Z system on the Zero. It’s a tiny printer and that would be a lot of hardware to squeeze into the small space. I do like the quad gantry bed leveling on the SV08, though.
My wish for a Zero Z axis upgrade would be a third lead screw. Two points form a line and the Zero bed tilts to the front and rear along the line connecting the two lead screws. I suppose Sovol engineers figured the bed was small enough that the tilt wouldn’t be too much of a problem but my Zero bed mesh sometimes has a low ridge along the front and a high ridge in the back, usually after inadvertently pressing on the front of the build plate when working on the print head. I can lift slightly on the front of the build plate and then the high ridge is in the front and the low ridge is in the rear. I chase that a few times before getting the bed level again. The stable points are at the extremes but once it’s level it tends to stay there as long as there are no forces on the bed other than printing.
PRO TIP: Manually move the nozzle to press on the front or rear of the bed in a precise controlled motion to level the bed without overshoot.
I wish the bed had three lead screws instead of two, and three bed leveling screws to manually level the bed, but with three lead screws with three Z motors (the Zero has one motor for both screws), the printer could do a bed mesh, then drive each of the three lead screws to level the bed without needing manual bed leveling screws, similar to quad gantry leveling.
After sanding the magnetic sheet under the build plate, when it’s level, it’s very flat and level.
You could quite easily give yourself limited manual levelling with silicone washers and longer bolts that attach the bed to the z rails. The washers will act as small bed springs and you tighten or loosen the bolts to move the axis. You’d get a few mm movement but on a 150 bed thats ample.
The reason for belted Z is nothing to do with levelling, its the precision and speed you gain by disregarding leadscrews, the more I think about it, its very much possible and I think we could do away with the leadscrew altogether and not replace them with rods.
That way the only space constraint will be under the unit, ill have to dive into the cad files but either have two separate stoppers slaved or see if there’s a chance to connect both sides via a rod
Lead screws mounted directly to the stepper motor are MORE PRECISE than anything involving a belt.
But the Zero has a belt on the Z axis.
It would be far simpler to do a direct drive screw (I’d do 3) setup than the 8 belts typical of a belt Z system.
No buddy, the zero has belt powered leadscrews, trying to say something has a belt when the final action is something different is just a poor understanding physics.
And there not more precise, they’re basic leadscrews, not ball screws. They’re prone to backlash, less precision, more friction and are tied to magic numbers if the pitch of the screws is not an interger of steps.
It’s a debate as old as 3d printing itself and tldr belted z is better as it removes so many variables from a system. Please remember, leadscrews were the cheap answer at the time to get the job done, but they add another layer of slop you cannot control well.
Think of it like this, a leadscrew is a gear in a system, you cannot re-print this, it is fixed, all math is constrained around the leadscrew, a belted system is fluid in that you can easily tune motor steps to linear motion and if done well is more precise.
Buddy please reread my post.
As for:
I’m well aware of the physics.
Toothed belts are NOT fluid and exhibit a systematic lead error as the trailing edge of every tooth slides down into the sprocket. https://www.sciencedirect.com/science/article/pii/S2590123022005199
I suggest you do a sensitivity analysis of the concentricity of the reduction sprocket typically used in a belted Z system. Then measure a injection molded gear you plan on using.
Anti-backlash nuts are VERY effective in overcoming backlash.
The point I was making was:
Packaging 3 direct drive lead screws in the base of a Zero would be 10% of the work of doing a belted Z and yield 90% of the benefit of an “ideal” belted Z system.
