Line width setting has no affect

[kmarts 0]

Hi all,

 

Started off with an anet A8 then built a metal frame for it. Recently changed out the controller for a RAMPS 1.4 and have been running through calibration and have an issue.

 

Now truth be told I've never tried to calibrate flow before but that's where I'm at. 

 

I have a solid 20mm cube that I set to 0 infill with a single wall with a width of .5mm, no top layer and spiralize outer contour enabled as well. What gets printed has a wall width of .63mm. When I set line width to .4mm I also get .63mm so it doesnt appear this setting is having an affect? Unless I'm not doing this right... 

 

If set at .4 and I'm getting .63 width the flow has to be reduced to like %63 which seems way to low. Curiously if I use slic3r to make the g code the line width ends up at .48 so I'm inclined to think this is related to cura but I dont know enough about the application.

 

Any suggestions?

[kmarts 0]

Here's the 3mf that should have all my settings if it helps

PI3_20mm plain cube.3mf

[burtoogle 513]

I sliced your project using 0.4 and 0.5 line widths and the extrusion amounts for the wall line scaled as expected.

 

Here's the 0.4mm line width extruding at 0.050 mm/mm rate (which is correct if you do the calculation)

 

 

And here's the 0.5mm line width extruding at 0.062 mm/mm (which is also correct)

 

 

I don't think Cura is causing the problem you observe. Hope this helps.

 

[kmarts 0]

1 hour ago, smartavionics said:

I sliced your project using 0.4 and 0.5 line widths and the extrusion amounts for the wall line scaled as expected.

 

Here's the 0.4mm line width extruding at 0.050 mm/mm rate (which is correct if you do the calculation)

 

 

And here's the 0.5mm line width extruding at 0.062 mm/mm (which is also correct)

 

 

I don't think Cura is causing the problem you observe. Hope this helps.

 

 

Please Pardon my ignorance here because I dont know a lot about this...

 

When you say 0.4mm line width extruding at 0.050 mm/mm what exactly does that mean and what is the calculation?

 

And if you're saying that it all looks ok, why then does changing the line width and wall thickness result in the same wall thickness between prints? I even changed my nozzle with a new one to make sure what I was using wasnt worn out or something. E steps per mm us also calibrated.

 

Doing the same in slic3r actually has an effect on the wall thickness with the same object. 

[burtoogle 513]

5 minutes ago, kmarts said:

When you say 0.4mm line width extruding at 0.050 mm/mm what exactly does that mean and what is the calculation?

 

The slicer assumes that the lines are rectangular in cross section (of course, this isn't true in reality but, then again, FFF printing is all one big approximation anyway). So for a line width of 0.4mm and a layer height of 0.3mm, 1mm of line needs 1 x 0.4 x 0.3 mm of filament which is 0.12mm^3. The filament dia is 1.75mm so that gives a cross section area of approx 2.4mm^2 so we need to extrude 0.12/2.4 = 0.05mm of filament for each mm of wall line length.

 

10 minutes ago, kmarts said:

And if you're saying that it all looks ok, why then does changing the line width and wall thickness result in the same wall thickness between prints?

 

Reality and intention diverge? Sorry, haven't a clue.

 

12 minutes ago, kmarts said:

Doing the same in slic3r actually has an effect on the wall thickness with the same object. 

 

I recommend that you use slic3r, then.

 

[kmarts 0]

4 hours ago, smartavionics said:

So for a line width of 0.4mm and a layer height of 0.3mm

But my layer height is set to .2mm So when you sliced it it was with different settings?

 

4 hours ago, smartavionics said:

I recommend that you use slic3r, then.

 

I wasn't trying to talk up another product here or put down cura. I'm asking in this forum because i use Cura and would like to continue to do so. Using other software does what it's supposed to do ruling out the printer which again is why I'm here.

 

If my line width is set to .4mm and a single wall setting with .4mm wall thickness should i really have to under-extrude by almost 40% to get that? Further to that if i set my wall thickness to .5mm and at %100 flow I'm still getting the same wall thickness as a .4mm setting then i have to change my flow % for the same filament depending what my wall thickness is set to.

 

I'm sorry but that doesn't seem right at all. I will try some other filament to see the results but i expect it to be more or less the same.

 

I'm just looking for confirmation that the line width setting in cura 4.0.0 actually has a real effect on extruder output because i can change that all day long and it doesn't appear to make a difference.

[burtoogle 513]

1 hour ago, kmarts said:

But my layer height is set to .2mm So when you sliced it it was with different settings?

 

Ah, I just noticed that your file doesn't contain any settings so I was using my settings. You should do File -> Save not File -> Export. This confuses Cura users new and old but the Cura devs do not seem very keen on making it obvious that to save a project file you need to save rather than export.

 

[kmarts 0]

Yeah this is all new to me... had no idea the settings wouldn't be exported.

 

I will do that tonight and re-upload. Maybe i have something set wrong.

[kmarts 0]

Sorry nothing last night... Kids had things to get to.

 

Anyways it appears I'm ready to eat crow on this one.

 

I can't explain why Slic3r extruded closer to the set line thickness than cura. However, when i finally got home last night i ran the same print with a different PLA.

 

Up until then i was printing MG chemicals white PLA. I put in some AMZ3D blue and ran the same print on %100 flow .5mm wall thickness. Object printed with .56 wall thickness. After dialing back the flow to %85 i sliced it twice, one with 85% flow .5mm thickness and the other %85 flow with .4mm thickness.

 

.5mm file printed with .51 wall thickness, .4mm file printed with .41mm wall thickness. So at first glance it would appear that my initial problems could just be from the filament?

[burtoogle 513]

Great, glad you have found a solution.