Z Wobble with 2 Independantly Driven Z Motors
-
Ormerod 1 (printed in red, like the RS logo) was an RS exclusive. I think we made 500 of them, for Christmas 2013. We updated parts and put shielded wire and housing on the Duet for Ormerod 2, which came out a few months later. Sold lots of them over a couple of years, but can't remember exactly how many!
Ian
-
Ok, I did have a good plan for this but alas the v0.6 and v0.8.5 boards do not support changing the microstepping mode. The A4982 drivers do, through MS1 and MS2, but on both of the first generation boards these are pulled up to 3.3V, or if you really want to you can cut the jumper and permanently drop them to full, half, or quarter step.
Plan B is hard wiring wiring the motors together. Thankfully I have a board kicking about that can do that (5V regulator board for a v0.6) without too much fuss. This will obviously permanently disable the G32 axis levelling. However, if this shows that the motors being out of sync can cause z wobble issues then I would consider the v0.6 & v0.8.5 boards of being in capable of x axis levelling.
Edit:
My plan was to create a macro with the following:- Home all axis.
- Move to 30mm above dead centre.
- Change Z microstepping mode to full step with M350 Z1
- Set the current z position to 30 with G92 Z30
- Run G32 a number of times.
- Move to 30mm above dead centre.
- Change Z microstepping mode back to default.
- Set the current z position to 30 with G92 Z30
- ...continue.
I would be interested to hear how someone with a Duet2 gets on with that. To be complete I guess the bed should be levelled to be good enough to not need bed levelling, but if that was a contributing factor I would expect the banding not to be parallel to the xy plane.
I will report on the effect of hard wired z soon.
Edit 2: Hard wired hasn't eliminated the issue. Will be comparing bed correction to no bed correction before changing back to independent steppers.
-
I'm returning my system to dual drive. There didn't appear to be any positive effect at this stage from running the steppers in series. Only negative; I needed to slow the system down so that it didn't lose steps on the Z.
I'm not saying it has no effect, it's just something else is having a far greater effect. Will try and set my z axis up again!
-
@DocTrucker It is not clear from your pic, but do you constrain the leadscrews at the top of the printer?
If so, don't do this, as this certainly causes Z-wobble. -
Yeah, tried both ways and over contraint was worse. On my sulk off to the shops I realised I could of made a mistake on the CAD in the location of the centre of the lead screw relative to the smooth rod. Will also try flipping the z nuts so the bodies aren't constrained by the x axis end blocks...
-
Looks like it may have been a noddy mistake! My attempts at de coupling the z axis by winding the nuts back did nothing because the nuts were a close fit in the x axis end blocks. Flipping the nuts made them much looser and the right hand rod wobbled around while the axis was going up and down. Think bought 6 trapizoidal lead screws and 2 of them so far have been wobbly!
This is the most clearly wrong thing I have found so far, so I am hopeful for the current benchy!
-
Time for Z-isolators, then...
Ian
-
I did a quick test with my current set up - T8 4mm leadscrews, Z isolators, not constrained at top, independent Z motor drivers, no switching to full steps - and I can't discern any difference between that and when I was using M5 (0.5mm lead) stainless rod with Z motors in series. So I think any further Z wobble on my setup is down to extrusion variation and using old filament. Food dehydrator has been ordered!
Ian
-
@droftarts said in Z Wobble with 2 Independantly Driven Z Motors:
Prusa i2 machines had a problem that the distance between the Z leadscrew and Z smooth rod were different between the X ends and the Z motor mounts, which caused the Z leadscrew to run eccentrically,
+10 points!As far as I can see the motor and rod holes on the z motor mounts are 17mm apart and 13mm from the edge of the part and on the top plate 17mm apart but 12.7mm from the top plate. This makes a lot of sense.I think I had made improvements with mounting the motor on a flexi mount. I'll now look to print a z motor mount and correct the spacing after I've double checked the main back plate CAD.You would have thought they would have fixed that when copying the designs to Steel wouldn't you?!! -
Hi,
I recently reworked by DBOT to try out bed leveling via a 3 lead screw setup.
Having some problems in the past with lead screws driven directly from the stepper motor shaft on this printer I mounted the lead screws using common lead screw bearings and connected the motors via belts.
Very happy with the result.
Frederick
-
Meh. False alarm. On careful remeasurement of the DXF files the base plate for the motors is wider than the main frame, giving the impression the rods are 0.3mm too tight when checked with the vernier. Investigation continues, but a z motor mount that allows me to adjust the centre of the z motor a little can't be a bad thing. Something is positioned wrong, just can't see what yet!