I have read your thread about backlash compensation and this thread.- I think the machine works for most people out of the box with the factory settings of backlash which is 0, i guess?
I guess the fast moving in this print produces failures at the weakest points which are the x and one y linear modules.
Whats your settings @Ev0lv3?
I believe he used G-code generated by brent and had the same results.
@MooseJuice The gcode I modified only had extrusion removed, was just curious if the movement alone caused it, and it sounds like the answer is yes.
@xchrisd Unless @Ev0lv3 comes back and says “oh yea I was manually enabling backlash compensation before every print” I don’t actually think that’s the case here, but it’s possible, so I mentioned it.
Unimportant side discussion:
As to why backlash causes these issues, I think there’s a flaw in the Marlin core. Yes, it comes from the factory disabled, but if it were enabled with a large value it seems to cause drift. In my opinion backlash compensation should be enabled - it’s extremely common in milling and other CNC work to compensate and work around gear slop, and that applies here as well. It’s a free performance boost. That being said the factory spec is 0.02mm backlash, which doesn’t really need compensation. In my case the X axis loosened up and enabling backlash compensation was a way to work around the slop without having to disassemble it
I was going to start looking at backlash compensation next, I am amazed at how well this double threaded screw print came out but the lines bled through some of the threads to the outside and its the only thing left i havent really tried to fiddle with…
I had a similar y-axis skew problem some time ago with my Snapmaker 1. It turned out to result from slippage in the shaft coupler connecting the y-axis stepper motor and the y-axis threaded rod. Tightening the set screws in the coupler solved the problem for me.