I recently purchased an Anycubic Kossel Linear Plus delta printer, and I'm interested in trying to print slightly below the 0.1 mm stated lower limit of layer height. I'm not interested in trying to push much below 0.06 mm, more just 0.08 mm to give some better surface roughness on critical parts.

I've found several articles that suggest that going below 0.1 mm requires lowering the printing speed, but the top answer on this question suggests that there are flow rates to consider in order to prevent jamming.

I'm referring to PLA in particular as I have read that the filament viscosity is also a factor, with ABS differing from PLA.

I'm interested in some benchmark settings, preferably for 0.08 mm printing. What is the ballpark range my print speed should be in to successfully print at that height?

  • $\begingroup$ Why not try it by printing calibration cubes or calibration towers? Do note that temperature is also a parameter to play with. $\endgroup$
    – 0scar
    Commented Sep 11, 2018 at 21:24
  • 2
    $\begingroup$ @0scar experimentation is always good, but that's not what I'm here for - I was looking for some empirical data, from those who have run more prints than I have so my first print on this Delta would be a bit more than "winging it". Nonetheless, I'm running the printer on a 25mm calibration cube, 0.08 mm layer height, slowed down to 45 mm/s on print speed at 200°C. I will post an answer with the results (will test with calipers) when it completes in ~90 minutes. $\endgroup$
    – ecfedele
    Commented Sep 12, 2018 at 2:55

1 Answer 1


To answer this question for anyone who will find this of use, I printed this calibration cube at 25mm to a side (125% scale in Cura) with my Kossel Plus using the following settings:

  • 0.08 mm layer height
  • 45 mm/s print speed
  • 200°C extruder temperature with PLA

Based on articles I read, I wanted to play it safe with the speed, staying well below the printer's rated speed of 60 mm/s, but still keeping the speed reasonably high to avoid any nozzle flow issues as describe in the linked question. The results were pretty good for a first test; the deviations, as measured with calipers:

  • X: -0.09 mm
  • Y: +0.38 mm
  • Z: +0.11 mm

This makes the average deviation across all axes about 0.13 mm, which is better than what I was hoping for, as that is about 5/3 of a layer's thickness. Whether or not running slower (or faster) or modifying temperature will result in better accuracy, I cannot be sure - but in this case, running the speed conservatively yielded positive results.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .