This question asks about computing the fundamental layer height unit for full Z-motor steps, on the basis of a theory that layer thickness can be inconsistent if the height is at different micro-stepping positions on different layers. The specific printer mentioned in the question is an Ender 3, which I also have, and while I've wondered in the past based on this video whether using whole-step layer heights helps, I haven't seen conclusive evidence that it really does.

According to the research in How Accurate Is Microstepping Really?, the A4988 stepper drivers (which this printer uses) have near-linear microstepping response, at least under the test conditions, which suggests there's little or no benefit to using whole-step layer heights.

Has anyone done experiments to determine/demonstrate whether use of layer heights that aren't whole steps actually leads to extrusion/layer adhesion problems or other structural or cosmetic issues?

I did a quick test with a 20 mm cube and 0.2 mm layers (5 full steps, left) and 0.1975 mm layers (1 microstep short of 5 full steps, right) and there is no readily apparent difference:

test prints

And repeated with single walls:

test prints

  • $\begingroup$ Could you report the printed height of both (using a caliper)? Just out of interest and for information as the 0.1975 mm layer cube requires "101.2" (effectively 101) layers to reach 20 mm, while the 0.2 mm needs 100 layers. E.g. it could be that it is doing 0.2 mm layer heights for many layers. $\endgroup$ – 0scar Sep 10 '19 at 7:11
  • $\begingroup$ @0scar: My caliper only measures to 0.1 mm so it probably won't be very helpful. I don't see how it could be doing 0.2 mm for all layers though - the Z moves are in absolute coordinates, and clearly right in the gcode. $\endgroup$ – R.. GitHub STOP HELPING ICE Sep 10 '19 at 10:55
  • $\begingroup$ You are right, not for all layers, but it must be some, e.g. those close to the full step. $\endgroup$ – 0scar Sep 10 '19 at 11:20
  • $\begingroup$ I'm also not sure where the idea comes from with the "full steps as layer height". You will rarely end up at exactly a full step at Z=0, so even "even" layer heights like 0.2 mm would have to stepper driver move to stepper to locations which are off by a few microsteps. Now, this might mean the stepper reliably falls back to the last full step, and all subsequent layers are identical again (except for the 1st), but that would just mean you've got far too low Z axis currents / too much weight to begin with. $\endgroup$ – towe Sep 10 '19 at 11:52
  • $\begingroup$ @0scar: Both measure 20.0 mm on one side, 19.8 mm on the other side; seems to be from damage to bed surface making bottom not entirely level. $\endgroup$ – R.. GitHub STOP HELPING ICE Sep 10 '19 at 11:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.