Taz 6 1st layer slightly too thick.

I have been having an issue with narrow protrusions of parts not sticking very well. I have noticed that the fill on these protrusions is not making contact with the table (round cross section, no flat spot to show contact).
I have increased the 1st layer line width to 135%, and this has helped, but the thin protrusions still show significantly less contact than the broad regions on the 1st layer.
I had printed a part with a brim, and decided to measure the 1st layer height.
the 1st layer is set to 0.425mm, but the brim measured 0.4826mm.
Is there a way to adjust the calibration?

What’s the Z measurement on a calibration cube? That’s probably a better gauge of the Z accuracy. Or design your own calibration print for fitment accuracy (ie, solid cylinder, and hollow cylinder). Here’s one from Thingiverse.

Personally, I think .05mm isn’t enough to worry about. It could be many things… overextrusion, probe disc or nozzle machining tolerance, filament color (additives in certain colors can cause dimensinal inaccuracies.), caliper accuracy…

Here are a few options you can try increase accuracy:

  • Reduce flow rate by 2 - 5%. This will cure the possible overextrusion. Ultimately, double check your e-step calibration (mark filament 120mm above idler, extrude 100mm, measure distance to mark… if not 20mm, calculate new e-step found in link above).
  • Adjust Z-offset through the LCD panel.
  • There’s also a gcode command to offset the Z axes… it might be M206 Z###. My machine is in the middle of a print, but a M501 should show the current setting. Make sure to issue a M500 to store the new M206 value.

I found the z-Axis adjustment in the LCD menu, and adjusted for the error. I also set the first layer line width back to 125%. I am getting much more uniform 1st layers now instead of starved sections where the stroke of the fill is short.
I did a factory reset, and found that it sets the Z offset to -1.5mm, the same as the thickness of the auto-leveling washers that hold the glass in place. This is different from the number that was initially in memory. That was something like -1.25mm (I don’t remember exactly, and I did not write it down).