I guess technically it is slightly racked, but only off by 8 or 9mm. That should be close enough right?
I got through the 3x3 calibration once with the current frame setup by using:
Machine width: 2036
Machine height: 1345 (larger than the measured height)
Calibration width: 500
Calibration height: 300
It got through the 3x3 calibration with a .48 fitness, but after the 5x5 calibration it returned a .16 fitness.
I spent lots of time trying different measurements with no luck. I ended up going back and trying the calibration again with the same settings (2036 x 1345) and instantly got the emergency stop after the 1st waypoint.
I will take a look at the onshape doc and keep plugging away at trying various other measurements.
Reading through the thread, I was wondering if you were confusing workpiece dimensions with frame (anchor to anchor) dimensions, but seeing your measurement pictures, that’s not it.
I’m wondering about the push-to-extend bug.
You might try increasing the increment you are adding to/subtracting your frame size. 1 mm is awfully small. Try something more like 20mm.I did not encounter this problem with my setup, so I’m not speaking from experience here, though. Someone correct me if I’m wrong.
If you can capture the output (save serial) and post it we can play with it in the calculator
When I took your original post’s CLBM line into https://m4cal.etskinner.com/ and gave it a 9999x9999 frame size (in other words just have it guess the anchors) it came up with these locations for the anchor points:
Just as an experiment (not sure this will work at all) Could you try entering in 1725 x 1725 for your frame size and then see if it can jog (just set the values, save, then reboot and do the retract extend, hang, apply tension) then see if it can jog at all? If it does, then I’d try a calibration with those numbers and see if it gets further.
also, maybe set the frame size to something very large and do retract / extend and just pull the belts out as far as possible (don’t go too far or they start spooling the wrong way – don’t ask me how I know ) , then retract all again and make sure they are properly spooled on?
Thanks for that. I’ll enter the values from there and try things out. Unfortunately I am just leaving town, so I won’t know how things will work out until next week.
Thanks for that. I’ll enter the values from there and try things out.
Unfortunately I am just leaving town, so I won’t know how things will work out
until next week.
not a rush. this is theoretical work, we haven’t had people test it and see how
well it works yet. But since you are having problems with the automated
calibration, this may get you something close enough to move on to real work
After everyone’s help and advise, I have managed a successful calibration! With a fitness level of 1.17 to boot!
I am very embarrassed to admit my mistake, but I can handle the abuse I turned my maslow 90 degrees at some point without realizing it and was trying to run calibration vertically instead of horizontally.
The frustration trying to figure this all out taught me alot about the machine.
great to hear (and I think that’s the highest fitness I’ve seen), could you do a
summary post detailing your frame, etc to make it easy for others to get a ‘this
works’ summary?
) , then retract all again and make sure they are properly spooled on?
I turned my maslow 90 degrees at some point without realizing it and was trying to run calibration vertically instead of horizontally.