regular pattern to layer sizes

For a model with vertical sides my prints show that the layers have different sizes, according to a regular pattern.

This shows up when I use aligned seam arrangement so that layers should be exactly the same.

In the attached file for model 'deer' there are 15 layers of 0.2mm layer height giving a design height of 3.0mm. Actual height is 3.01mm.

The pattern 3 big layers(b), followed by 1 smaller layer (s). The 15 layers are 3b 1s 3b 1s 3b 1s 3b.

This same pattern is observed in other models, big and small, with taller vertical sides, 0.2mm layer height.

I use slic3er software ver. 1.2.9

Printer is Prusa i3 rework, with bowden extruder

This phenomenon is a great problem when it comes to making vertical holes for bolts and nuts.

Any one suggest what is the matter; what further information could I provide?

Edited 1 time(s). Last edit at 09/29/2018 10:56AM by Davedave.

I wonder if it's a slicer problem or caused by your Z-axis? Is it a matter of microstepping and mismatched layer height? Have you tried to increase stepper current?

Thanks! However the layer height seems ok, since 15 layers works out at 3.0mm and it printed at 3.01mm. The error is in the X and Y directions, leading to larger and smaller layer areas.

I, too suspect Slic3r, so I'll read up about microstepping, adjust potentiometers and give it a try.

Edited 1 time(s). Last edit at 09/30/2018 06:21AM by Davedave.

Do you have adaptive slicing enabled in print settings ?

Since you suspect the slicer, did you check if the problem is visible in the generated gcode?

It sure looks like a very common Z axis problem. Does it do that if you slice with Cura? Does it do it on taller prints?

It looks like a cylical error. That's going to add up to nearly the correct height because the average of the layer thicknesses will be the layer thickness you used in slic3r.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Thanks for all the input.
I've drawn a 20 x 20 x20mm cube to use as a test model.

I have updated Slic3r to 1.3.0 dev, and checked the G-code manually ( A bit of a long job, but using nraynaud.github.io[/u] website made it a bit easier; can anyone suggest a more friendly site?).

Also, is it possible to update Slic3r to a fully released version of 1.3.0?

In Slic3r settings adaptive slicing, I note that the profile sketch of layer heights vs model height had a few very minute wobbles on a regularish frequency. I managed to straighten these out. Also I note that a huge chunk of the profile, i.e. the lines representing individual layers (approx 20%) was missing!

I have yet to print, since I am still reading up.

Once again, thanks