I built the Burly MPCNC over a year ago and have enjoyed the additional Maker equipment, but experienced problems with the Z-axis losing steps. Many others seem to be having this problem as well. Researching the forums & websites suggested checking the hardware, adjusting the feedrate, adjusting the acceleration, increasing the stepper current. None of these steps eliminated the problem. Yesterday the machine was running a basic carving design and kept losing z steps and would plunge deeper and deeper. SOUND FAMILAR??? Running an aggressive up-down-up-down gcode routine on only the z axis could not recreate the problem, at which point all of a sudden the problem became clear. IT ONLY HAPPENS WITH THE SPINDLE RUNNING. The spindle is a BLDC unit and the three power leads are in the same cable track as the z axis stepper leads. All this time and the root of the problem was electrical noise induced into the stepper leads from the spindle leads. After installing a shielded cable for the spindle the problem has been 100% resolved. The foil shield of the cable has been connected to the common ground for the whole MPCNC. It should have been obvious, but the thought that induced noise could upset the relatively high currents from the stepper drivers seemed unlikely. So for anyone out there that is experiencing lost Z-axis steps check to see if the problem only occurs with the spindle running and if so then consider a shielded cable to the spindle. “I CANNOT BELIEVE ALL I NEEDED WAS SHIELDED CABLE TO THE SPINDLE!”
1 Like
Ordinarily I would be skeptical, since like you said it seems unlikely that the induced noise could have a significant effect on the high-current stepper wires. But it sounds like you’ve done your homework and the up-down cycling without the spindle points conclusively to it being spindle related. By itself that wouldn’t rule out, e.g. if there were a weak shared power supply that is under-powering the stepper drivers when the spindle is on, but shielding wouldn’t fix that.
So as unlikely as it seems, I think you have ruled out everything else. That’s good to know that it is possible for interference to cause missed steps.