Extrusion Multiplier and Extrusion Width

[Mount PrintMore]

I've been fiddling with calibrating my machine again (new roll of material from a new supplier).

I had read some discussion of folks printing an object that had a single extrusion width thick wall and measuring it. It sounded interesting, so I thought I'd give it a try. I attached my factory file (I use Simplify 3d). (edit, it won't let me attach a factory file or gcode file). If anyone wants a look at it, send me a PM and I'll email it to you.

I ran the same part with the extrusion width set to auto (0.42 mm) and various multipliers ranging from 0.9 to 1.02.

I cut these up with sharp scissors, then used a micrometer (not calipers) to measure the thickness of the walls. I measured two walls of each part.

All of them seem to be a bit wide.

EM 90 wall 1 0.696, wall 2 0.771
EM 94 wall 1 0.849 wall 2 0.67
EM 98 wall 1 0.906 wall 2 0.919
EM 102 wall 1 0.975 wall 2 0.977

I was really confused by the variation in the first two test pieces. All I can say is that in my fingers the two walls "feel" different too, as a sanity check on my measurements. I measured each wall 4 times and reported an average. I stayed away from the first layer or two as it is a bit fat from being squished on the bed. I also tried to stay away from the standard surface defects.

I find these results hard to believe, so I went through all the settings and am rerunning the parts now.

[insta]

Measure the roll multiple times over several feet, and for each measurement take two and rotate the second by 90 degrees (to catch ovality). You could just have a bum roll of plastic.

[Mount PrintMore]

that's what's odd about it. I measured the roll when I received it. If anything, it seems to be a bit undersized.

I reran the parts and got similar results. I didn't bother cutting them up and using the mic. I just checked it quick and dirty with calipers.

The inconsistency in the parts (4 walls of the same part) bothers me too. It's like the slicer is doing something odd with the geometry.

[insta]

If you want to attach .fff files, zip them up first. The forum will accept ZIP files.

There's calibration steps I posted elsewhere to get your esteps dialed in if they've drifted for one reason or another. Not sure where they ran off to though...

[jsc]

How are you generating the test object? Don't bother trying to model a thin wall cube. Just model a cube and set infill to 0% and no top and bottom layers.

[Mount PrintMore]

Here's the files.

I did model the thin wall. Now that you mention it, I can imagine why you wouldn't want to do it the way I did it.

[Mount PrintMore]

On the thought that the slicer was behaving badly, I modelled a part with a wall that was two passes thick (0.86 mm to leave some room for tolerance).

I printed the part with the auto extrusion width set and an extrusion multiplier of 0.9, and low and behold I got a nearly perfect part. I cut it up and measured the thicknesses with the micrometer. The average of four measurements was 0.867 and 0.863 mm respectively.

When I cut the part up, it was clear that the two passes of the extruder had fused together well, but on the top of the part you could clearly see a seam where the two passes had occurred.

I've attached the files.

[Mount PrintMore]

I have one last comment on this, then I'll be done with it here.

I reran this with a part sliced as recommended (a solid part with 0 top and bottom layers with 0 infill). Indeed if I run with a single perimeter or with two perimeters, I get the expected results (the measured wall thickness is very near the nominal value).

However, there is one final quirk with Simplify3D, that I think I'll go post on their forums. If I take this part, and put 4 of them on the build plate and create four different processes for them the only difference being the extrusion multiplier, only one of the four parts prints correctly (the part defined by the first process).

This is something I've noticed before (in nearly every version of S3D that I've used), but didn't figure out until today. I could never really sort out what I was seeing due to the other issues with the slicing.

[jimc]

so with the other 3 models, when you say they dont print correctly what do you mean? what differences are you seeing?

[jsc]

When you Prepare to Print, you should get a dialog popup that is titled "Select Processes for Preparation". At that point, you should Select All, Continuous printing.