I’m starting to calibrate my LR4. The gantry is wide enough for a 4ft sheet. I tried a shallow cut with a v-grooving tool. Should have measured 1100mm along the X axis, but came out at 1095.5. I know I can correct this in the firmware but I wanted to check whether that discrepancy was big enough that I should be e.g. tightening the X belt first.
Y axis came out a lot closer - 599mm when I asked for 600mm.
This may not be a popular opinion here, but you might be able to correct both the X and Y distances using belt tension adjustments only. This works best if you use the adjustment screws in your belt holder, rather than trying to move the belt holders to different screw holes.
Another thing that you might want to check is that both the Y belts (long axis, the ones with the rail and non-rail sides) need to be tensioned independently. On my LR3 build, there was about a 2mm difference in distance between the rail side and the non-rail side. (note that you are saying your X axis move was 1100mm, and your Y axis move was 600mm, which is opposite from most people’s builds)
I have M3 adjustment machine screws so that I can finely tune each belt, and the thread pitch of a 3mmm machine screw is 0.5mm. so 4 full turns on one of the screws was able to bring them to exactly the same distance. I was also able to get both Y belts to be within <0.5mm on a 2800mm move, and get the X belts to be within <0.5mm on a 1430mm move.
The BIG caveat is that you don’t want to overtighten the belts. They shouldn’t sound like a guitar when plucked (a poorly tuned bass, maybe)
If it’s a new machine, I’d check for mechanical issues first. Did you use locktite/thread locker on the grub screws that hold the pulleys to the motor shafts? If they’re not tight the pulley will slip and not move as far as it should, and even if they’re tight to begin with vibration may work them loose over time.
Minimum strength threadlocker (e.g. “blue” locktite) will prevent them from loosening by themselves but still allow them to be taken apart with the proper tool without needing to apply heat.
I would suggest taking measurements without cutting in order to validate settings. Put a piece of tape at X0Y0 and touch off with your V bit. Then move close to X max and do the same. Measure that distance. Do same with Y.
Yes, I’ve got a long X-axis and short Y - overall it’s about 4ft x 3ft capacity. This way I can scale up to do 8x4ft sheets if I need to, but I don’t have a massive unit taking up space while I’m only doing small jobs. The Y-axis move was consistent near Xmin and Xmax.
Interesting suggestion about tweaking the tension by a controlled amount. I’ll scale up the 1mm error over 600mm to whatever it needs to be over the whole length. I don’t have enough adjustment available to take 4.5mm out of the X axis though so will have a think about whether to shorten the belt a few millimetres.
I’ll buy some threadlock too. I’d be surprised if it’s the culprit here but I don’t want to write off a slab of wood later.
Well, I went out to the garage to adjust tension. Opted not to in the end as the belt tension is already as high as I’d like. Will adjust firmware instead, at least for the Y axis.
You should verify with a non-cut first, like the squaring test. Poke tiny holes with a v bit and measure again.
If the pokes are right it is a CAM or endmill issue. If the pokes are not right it is a firmware issue.