Finally starting to run really big (>1000×500) calibration grids today -
Last calibration was fitness 0.99749 - has been >2 on smaller grids.
fitness is turning out to be a poor estimate of accuracy. If you have few points
on a small grid, you will get good fitness because you don’t have enough of a
baseline to notice errors
My guess at a measure is ~ 93.75 in so 238.25 by 138.875 so 3527.42 The diagonals measured within 1/4 inch.
providign units here may make this make sense.
Most recent calibration:
tlX:-1, tlY:2403.3
trX: 3510.7, trY: 2401.8
brX: 3503.1, brY: 0
take a look at https://lang.hm/maslow/maslow4_frame.html and go for the biggest
grid that will stay in the green and not go so close to the edge that you will
fall off your material
also go for more points (especially as you go to a larger area)
I am noticing 2 arms don’t pull as tight with the same calibration limit of 1000 - I was running at 800 with significant sag. Is there a way to monitor motor current?
there is a setting in the maslow.yaml (configured via the UI) that enabled a
debug screen that shows motor currents.
Also looking for a good way to zero the bit- hard to see, so I’ve been lowering until interference on scrap.
David Lang