So 1.17 is calibrating fine, but one thing bugs me. My fitness is 1.5 to 2. Which I assume is pretty decent. But when I measure with a tape measure from the center of bottom left anchor to the center of top left anchor the length is about 1 inch off.
So I have been calibrating everyday, trying new things.
One thing that seems to have improved my anchor location to tape measurements is adjusting my belt end extension size… Assuming the stock part was 30mm end to center of anchor hole. I did the same originally… And mine are 40mm (3d printed), but these all resulted in my frame size calibrating a hair smaller then what the tape measure showed. I just got a 2.2 fitness and my BLy to TLy came the same length as my tape measure, by making my belt end extension 48mm. 48mm is the end of the anchor to the outside of the anchor hole, not center. Not sure why this works better?
Am I right in assuming that the fitness score in version 1.17 the higher the score the better? And my tape measure measures better.
I also discovered that on my frame, I did a successful calibration facing North. Then did a calibration facing south.
But then once the yaml was updated with my South facing anchor points… When I switched the machines orientation back to “north”, it would not calibrate because of the anchor positions existing in the Yaml were confusing it.and I got the over tightening belt situation in the corners. So I edited the yaml to my known ‘north’ facing anchor locations, and it would calibrate successfully again.
So I’m wondering if when I tried to run 1.23, I uploaded and Used @bar 's Yaml, and his yaml coordinates…and the calibration didn’t like the initial anchor point locations, and this is what causes the belt tension and snapping.
Worth noting, that changing the belt end extension size, never made the calibration more tight or loose, just changed the frame dimensions once completed … Not sure why it likes my 48mm vs 40mm better.
Next time I will try 1.23 and download bars yaml but update to my approximate dimensions before clicking find locAtions.. and maybe I’ll be able to calibrate with 1.23 new math.
Cheers