MakerGear Forum

When I use a new filament, I've been "calibrating" my extrusion settings by printing a small open cube (no top, bottom, or fill) with a single pass perimeter. I then measure that, and adjust my extrusion multiplier until I get the correct wall thickness. I have a MakerGear M2E, with a 0.35 mm nozzle. With 0.2 mm layer thickness, I'm shooting for a 0.4 mm width. I'm currently using eSun PLA, and the extrusion multiplier that gives me that wall thickness is 0.94.

The problem is that when I than go to actually print something, I get defects that are consistent with under extruding. I get tiny holes in the top layers where it is cross-hatching, and sometimes even on the bottom, despite some extra "squish" there to get good adhesion. I typically print with 3 or 4 perimeter passes, and again, on the top layer, you can see places where the perimeter extrusions are not fused together.

The test structure I'm working with is one I downloaded to test bridging. One thing that may be part of the issue is that the "top" of the print where I see the defects isn't the highest layer:
Is there a better way to do my calibration? I've been thinking that I could do a multiple pass perimeter on the same model & measure the combined thickness.

Is there some other setting that might be affecting things that I need to tweak? I've tried various speed settings, and it doesn't seem to make much difference.

Thanks!

Gwhite wrote:defects that are consistent with under extruding

Sounds about right ...

Once upon a time, a guy who knows about rheology walked me through what happens inside a 3D printer's hot end. The math describing the plastic flow requires highly nonlinear three-dimensional time-varying differential equations (which I straight-up do not understand), boiled down into the slicer's handful of linear coefficients setting speeds, temperatures, and flow rates. Slicing and printing work reasonably well, but what seem like simple problems generally do not have simple fixes because the algorithms don't model the actual behavior inside the hot end and onto the platform.

Edit: some of that discussion made me feel like I'd taken a direct hit from the Stupid stick ...
https://softsolder.com/2012/02/01/rever ... eleration/

The Extrusion Multiplier provides a simple linear knob for the extruded width, assuming you've set the correct filament and nozzle diameters. The thread width depends on many other factors, all of which the slicer completely ignores, and will vary as those factors change.

You can increase the Extrusion Multiplier (in steps of, say, 0.02) until the infill voids Go Away, which will increase the outside dimensions by a little bit, because the threads will be slightly too wide than the slicer expects:
https://softsolder.com/2013/09/17/extru ... s-thereof/



Basically, you can adjust the slicer's knobs to make any single part come out the way you want it. Those knob setting will not produce equivalent results on other parts, even though they have similar geometry, because the simple model doesn't match the different rheology.

The results will be close, which is the miracle of DIY 3D printing, but the fussier you are, the less you'll like them.

measure the combined thickness

That won't do quite what you want, because the slicer positions each thread to put its outward-facing wall at the correct location relative to the part's exterior. The total width error of multiple threads is therefore determined by the width error of one thread, because the outward side of the innermost thread will be at the right position.

When surface finish is important, high-build primer and paint will be be your friends...

When I first ran into this a few weeks ago, it was on a non-critical part, so I did just what you said & cranked up the extrusion multiplier. I guess I will just have to learn to live with tweaks like that.

I spent several days working on getting rid of (or at least significantly reducing) zits when printing with PETG. That stuff is seriously non-linear, and I finally declared a draw between what i was getting & what i had been hoping for...

3D printing keeps getting better, but it is still a LONG way from plug & play...