Extruder Calibration

Hey guys,

I finally completed the construction of my Prusa i2! I have managed to print some cool stuff in ABS including a Yoda head (that got lost), a 6-hour print that I designed myself (125x82.2x80mm print envelope) and a ton of incomplete prints of various stuff (stopped them myself to recalibrate GCode or thermistor disconnected, etc.). Print quality is passable, with quite a bit of blobbing around places and 0.3mm layer height that looks rather 'rounded' if you get what I mean.

However I have had trouble printing small objects such as the Tiny Planetary Gear Set, Thin-Wall test, or Brix' Adjustable Fan Mount. It always seems as if my extruder is feeding way too much plastic and causing it to be 'mushy' when a new layer of plastic is applied. I measured and re-measured my E-steps/mm and they are correct - 400.14 for 1.8º steppers, x8 microstepping, Greg's Accessible Extruder (which to me, based on what I've read on the wiki does seem rather high).

As a result of having been unable to print small items, I set out to try and properly calibrate extrusion to get, at the very least the 0.5mm test done at 0.3mm layer height. So far, I got the extrusion multiplier down to as low as 0.517 to try and get 0.5mm layer width, but I am still having problems with a massive column of blobs appearing at the Z layer height change. I tried reducing and increasing retraction as well as altering its speed (or a combination of the three) but to no avail. What may be causing my bad print quality?

I am using Marlin, Repetier 0.40 (45), Slic3r 0.9.1, Sanguinololu 1.3a electronics with genuine Pololu stepper drivers, J-Head hot end with 0.5mm nozzle and I am printing ABS at 125 ºC - HB, 238 ºC - E first layer, 113 ºC - HB, 235 ºC - E every layer afterwards.

Thank you in advance.

Blobs on retract are commonly caused by the extruder missing steps on the retract, usually because the retract speed is too high.
Put a mark on the big gear and watch it during the retract, make sure it returns to the same point it was at before the retract.

You can run the single wall test with no retract and I would suggest you do it this way because it eliminates an additional source of problems.

I assume you've measured the filament going into the extruder for a known extrusion 50mm in when 50mm requested?
There will be some farting around to get things perfect even if this is correct though IME <10%
The reason it's not just volumen in == volume out that the slicer has to approximate the cross section of the output, and it's volume out calculation is based on the approximation.

The one part of calibration that did my head in for the longest time was the thin walled test object, I would calibrate to .5mm thickness and then could not print a solid filled object. The clue with the thin walled object is to check the gcode to find out what wall thickness should be for a single wall, and then calibrate till your walls match.

Disabling retract and messing with extrusion multiplier gave me thin walled cubes that made me more hopeful that I'm on the right track than I was before.

I calibrated my extruder before by, as you, Polygonhell, said, feeding a known quantity and measuring how much is actually fed (in my case 100mm was the quantity).

I was thinking about the fact that 0.5mm wall width seems a little low for a 0.5mm extruder but what exactly am I looking for in the gcode? I am illiterate when it comes to gcode. So far a wall width of 0.8mmish gave me the most successful results but it's clear that the value for the corresponding extrusion multiplier is far from calibrated.

Thanks for the help, it is much appreciated.

If your using Slic3r the expected wall width is printed as a comment at the top of the GCode.