Welcome to the SeeMeCNC Community

Rather than highjack one of the other threads on this subject, I figured I'd start my own for what appears to be a very important, yet highly over-looked topic. I've spoken with NUMEROUS people who "can't figure out why they can't get quality prints" and found they have never performed this step...

So I'd like some clarification from the masters.

1. What I have done is marked points @ 10cm distance along my (white) filament.
2. Extrude the filament until one of my marks is exactly at the clear filament guide.
3. Extrude 10cm of filament and check to make sure I am sitting at exactly my next mark (and I am, after some tweaking of steps per/mm).

So I thought I was good. But then I read that I should be measuring the filament that came out of my extruder. The problem I have is when I extrude filament, it comes out all twisty curly, not in a straight line so there is no way I could measure that. On top of that, logically it doesn't make sense. If I extrude filament from a height of 100mm isn't it going to stretch under it's own weight as it comes down? Am I missing something?

the next step after setting your steps per mm correctly.

would be to calibrate your extruder in slicer software.

in slic3er this would be adjusting your extruder multiplier.

in kisslicer this would be adjusting your flow tweak.

i never measured how much filament was coming out of my hotend. my prints seem to come out pretty good.

Ok, thanks!

I thought that the steps per mm on the extruder was used as a flow control for dialing in the thickness of the extrusion coming out of the hotend instead of measuring the filament coming off the spool... Like in this thread: http://forum.seemecnc.com/viewtopic.php?f=54&t=1163

Is this correct, or have I missed a calibration step?

Steps per e is the number of stepper motor steps required to move a set distance. For instance, when looking at the beginning of the configuration file in the firmware sketches, you can see how the steps per (direction) are calculated. The number of physical steps is multiplied by the number of microsteps and divided by the circumference of the pulley attached to the motor. Using steps per e is not the way to control extrusion thickness. It affects the extrusion width, but simply adjusting steps per e to get a thicker line is a haphazard means of doing so. Steps per e, when dialed in, should not be changed. Now, calculating your steps per e is a bit of an art form in my opinion. But the starting point is correctly measuring your filament diameter (measure it in a couple directions and in a couple of spots). Then my preferred method is to find a print that has a flat surface several mm above the printbed. The finish of the surface should be such that you have little to no ridges or dips in the surface between printed lines. --v--v--v-- = under extruded --^--^-- = over extruded (Looking in profile at the top of the part with the --'s being your printed lines). Reason why this can be difficult to determine is you have to print warm enough and slow enough that the hotend has no trouble depositing the correct amount plastic and the printers movements need to be correct. If you are seeing something looking like this || || || where the lines are not evenly spaced, you know that you have a problem surrounding the printer's movement and trying to calibrate your steps per e needs to be put on the back burner until that's corrected. Also, it's important that your extruder does not have any slippage in driving the filament through the bowden tube, else we again, can't accurately calibrate steps per e.

Reason why this is important: When a slicer is calculating the amount of plastic to lay out, it looks at the diameter of the filament and determines the length necessary to produce the desired volume. Volume in needs to equal volume out. Because of how PLA is extruded onto spools, this is essentially 1:1, however ABS is spooled with essentially small "voids" in the filament producing a filament that is about 85-90% "plastic". That's why flow compensation is used. When your steps per e and filament diameter are correctly set, changing extrusion width/nozzle size should have almost no impact on print fill quality, speed changes shouldn't affect the fill quality (with corresponding temp changes) as the slicer doesn't care about the individual parameters as much as it knows I need "X" much plastic to fill "Y" amount of space, so I need to push "Z" length of "D" diameter filament to the hotend.

http://richrap.blogspot.com/2012/01/sli ... s-and.html Rich does a great job with explaining the math based upon the results of measurements in this blog post...