I have a question about @Simplify3D and extrusion widths – I thought I understood, but clearly I don’t.
I was looking at @SPANNERHANDS_3D_Prin 's Prusa i3 S3D settings and he has his extrusion width set to 0.43 (rather than my current setting of Auto (0.48?)).
So, I thought I’d try a print with the extrusion width set to 0.43 – nothing else different. And it under-extruded, very obviously. Some small top surfaces have slight gaps, multiple perimeters sit next to each other but don’t quite touch, and – the most obvious – the brim consisted of separate rings. not quite touching (so useless as a brim).
No pictures as the part is copyright (for a client – I’ve reprinted with my usual settings for him).
Now, @SPANNERHANDS_3D_Prin has an extrusion multiplier much higher than mine (1, whereas mine is 0.95), and upping my extrusion multiplier may well fix things, but:
For the brim (for example), surely S3D is supposed to place the concentric rings of the brim according to the extrusion width? And surely that’s the case for the rest of the print as well? Surely changing the perimeter width shouldn’t mean we have to tweak the flow rate?
Remember, all I have changed between the (absolutely fine) print with my default settings and the print that looks obviously under-nourished, is the extrusion width, nothing else.
Can anyone shed any light on this? I’d love to know what’s going on here as I thought I understood, but clearly I don’t.
@Baked_Bean Thanks – that’s mostly about having a perimeter width that fits nicely in the available gaps in the model. I get that, but I don’t get why, for example, the brim is fragmented, nor why the eventual print looks so starved.
I think it would help to make a test piece so you can show us. One thing to consider is whether the nozzle orifice really is 0.4. It might be a case of the plastic stretching too thin because of that. Another is that the Z=0 height or flow rate might not be really set right and a wider bead might have been hiding that.
@Baked_Bean Slic3r and Simplify3D do their volume calculations differently, that link is irrelevant.
S3D uses “extrusion width” as the actual pitch between adjacent strand paths, and calculates volume as the ideal rectangular shape, equal to WidthHeightSpeed and then multiplies that times ExtrusionMultiplier as a calibration factor. You tune extrusion multiplier tuning to make sure the strand overlap and void space between strands is correct, by printing a SOLID object and making sure it’s not over/under-extruded.
Slic3r uses “extrusion width” as the measured thickness of a single-wall box, calculates required volume from the theoretical oval shape produced by extruding a squashed noodle of filament, and then sets the actual pitch between adjacent strand paths closer together than the “width” to ensure strands overlap and bond properly.
It’s different calibration concepts and different algorithm methods of getting to what should end up being the same result.
@Mark_Wheadon What you might be seeing is a change in the shape of the extruded strand due to the different nozzle diameter / strand width ratio. You get a different strand shape when the extrusion width is less than nozzleDiameter+layerHeight. Micrographs I took with a 0.4mm nozzle, deliberately under-extruding to highlight the void spaces"
I have noticed this myself. Especially if I use a first layer extrusion width multiplier in simplify 3D. I set my S3D slicer at exactly 1.75 filament diameter, but calibrate every new roll with single and double wall EM method using a 0.5 Extrusion Width. Using that calibrated EM I get very dimensionaly accurate parts using S3D on that role. But S3D fails to get a good first layer if I use a 125% extrusion width for first layer (leaving gaps between like especially).
To me this feels like a bug in how it does first layer modifiers. So my solution is to stick with 100% first layer width and just modify first layer height to get a 0.3mm first layer thickness which tends to give good strong results will excellent adhesion.
But my EM does not work if using a different slicer program. I need to calibrate each slicer independently for EM.
@Eclsnowman I do the exact same thing – leave first layer width alone and use ~0.3mm first layer height. Never saw any real point in messing with first-layer width, it doesn’t seem to stick any better for me.
I do the same – leave first layer width alone but just make sure it’s a decent thickness (and perhaps a little hotter). So 0.15mm layer height gets first layer height of 134%, 0.1mm gets 200%.
I have tried a different approach. To get better adhesion I have put first layer height at 50% and increased the width to 105%. That seems to ensure that the nozzle is very close to the bed and there are very little gaps in first layer extrusion.
@Eclsnowman do you modify your first layer via the percentage adjuster for modifying the first layer (i.e.; if your printing 0.100 layers, you set the first one to 300%) or do you make a separate process for the first layer only?
