Mini auto-leveling details

I opted for a Mini for my first 3D printer and absolutely love it. What a remarkable piece of engineering. I’m only two weeks in, got a question about the auto-leveling details.

I think I read here in the forums it’s using ohmic sensing but, it touches off on each pad twice. Almost like it’s using the servo motor load for the rough approximation then going back at a much slower speed to fine tune. Is that what it’s thinking?

They aren’t servos, just regular, open-loop stepper motors. The reading is more accurate the slower the z-axis, but it would take all day for it to go slow from the top of the probe, so that’s why the double touch.

You might like this post:

My bad on the typo, I meant stepper. I’m more used to industrial CNC machine tools and sometimes mix vocabulary. Thanks for the interesting thread. Two thoughts on that.

M109 S140                    ; set to cleaning temp and wait
G1 Z150 E-30 F75             ; suck up XXmm of filament
M109 S140                    ; heat up rest of way

My start/end-gcode from the current Cura default PLA profile says this but that never made any sense to me. Near as I can tell it’s saying wait until you hit 140, retract then wait until you’re at 140. Those same g-codes in the thread are S160 and S170 which makes somewhat more sense.

you may want to clean your nozzle shortly after each print, or before it begins the homing process. If the bed visibly moves during the probing process the nozzle isn’t clean enough.

I’ve noticed my printer will sometimes visually deflect the bed on the fast -x-y and -x+y touch off but not on the other two or the slow touch on any of them. When that happens the z is way too low so I cancel and run it back through the cleaning cycle. Before I knew to watch for it etched the bed on one failed print.

Don’t quite get why it does that. If the nozzle was dirty and the filament material is acting as an insulator I’m thinking it’d deflect on all eight touch offs. I never see deflection on the +x fast or slow.

I have had some trouble with the leveling process deflecting the first washer and not the rest. In my case, it is caused by some plastic residue on the nozzle. I think the residue is squeezed out on the first washer so the electrical contact happens properly on the rest of the washers.

I primarily print PLA on my Mini. I have found it is most reliable during leveling if the hot end is at 140C throughout the process.

Clean up your hotend, replace or flip ths wiper pad and try again with a cleaning and probing temp of 140. That does it for me.

I think I have been cleaning and probing ABS at 160 without issue.


Cool, thanks!

I cleaned up those two washers and that seemed to help a lot. I couldn’t see anything on them but then again my vision is not exactly up to factory spec these days. I think this also might have something to do with PLA’s tendency to drip or string a bit during the auto-leveling procedure. Some things I’m finding that help:

  1. Warm up the hot end and extrude some material before the first print of the day. That way you’ve got the amount of filament the firmware expects, in the location within the extruder it expects.

  2. Fluff the wiper pad (a little, not a lot) with the supplied pointy thing before the first print of the day.

  3. Watch the auto-level sequence like a hawk. Any visible deflection, hit cancel then restart the cleaning cycle. It almost never ends well if you see deflection.

PLA is an awesome material, it seems just a little more fussy than some of the others.

If I didn’t mention, I absolutely love this printer.

If you are seeing dripping or stringing during the cleaning / leveling routine, your hot end temp is too high. I have used 3 or 4 different brands of PLA and none of them are runny or drippy at all at 140C. You should double check your wiping / probing temps in the start up gcode.

Make sure the set probing temp line in gcode is not warming the nozzle higher than 140C before probing begins.

M109 S140 ; set to probing temp

I’m using some eSUN PLA I picked up from Amazon and this spool definitely strings and drips a bit @ 140C. Next up is a spool of Lulzbot green direct from Lulzbot. I’ll be curious to see if there’s a difference.

Tonight it totally nailed the auto-level and bottom layer on the first try. Is it wise or advisable to save those offsets and is it possible to embed them into the startup gcode?

When it’s got it’s game on the auto-level usually results in about the same offsets. Unless you do maintenance on the mechanics or the environment changes significantly I’m thinking that’d be a good and reasonably reliable place to start from day to day. Split it out and do a leveling calibration only when you begin to see some drift or something in the mechanics or environment is known to have changed.

That is strange. I have some black Amazon esun pla. It is just soft at 140C. No chance of it dripping.