Xy calibration.. close but not there

Hello, just got my machine put together.

Been using the calibration test pattern from @RobertA_Rieke and got pretty close. I did the standard XY against ruler calibration (450mm moves) first, then ran the test pattern, then reran the ruler calibration and then ran the calibration file again. Belts @ 4lbs for 1 inch travel, Vwheels seem tight but that feels like the most likely cause of some minor issues. Measured width of the outlines, .125 in so bit width looks good. used easel default settings for mdf (40IPM)

One problem is that the errors from these small numbers don’t scale… one mistake i made first round of the calibration file was producing a new steps/mm directly from those errors… that made things way worse (10mm in 450mm).

anyway steps/mm from 2nd round of ruler
Y steps/mm: 39.873
X steps/mm: 39.866

stats from testing

print itself

anyways came here after doing it twice and not really coming up with a change.
previous steps/mm was 39.92 for both X/Y on the first ruler round… so actually reduced despite the stats.

maybe I need to do something different with calibration cut? I found these circles really hard to measure. if I’m pretty confident things are “mostly” round, just do some things easier to measure for repeatability