Fresh calibration, everything looks good. Corners are robust and I measured the frame very precisely. I retract, extend, mount the Maslow, and tighten the belts – they feel great and taut. Then if I move the Z axis, or any axis really, the belts loosen up, especially BL and TR. Same thing happens if I start a cut, the first thing it does is loosen the belts. I can go back and have the Maslow tighten the belts again, but same issue if anything moves.
Is this normal? There’s a lot of play in the machine. Here’s an example of the BL belt (tried to aim it down what should be the centerline of the belt):
During calibration the machine should set the ZStop automatically. Are you seeing the message about the Z stop during the calibration process? It should say “setting Z stop” or something similar after one of the pauses where it calculates. If not you can set it manually and you may need to upgrade firmware
I bottomed the Z axis out before calibrating, but noticed when it was done the Zmin/Zm on the dashboard tab was like -34mm when it should be zero. Not sure if that’s causing the issue or not.
Didn’t save the calibration log so no idea if it said it was setting the Z stop, but I can save the log next time.
Did another calibration made sure the Z was all the way down and still running into the same issue. Kinda stumped here. I bought it second hand, maybe the PO manually put in some weird settings?
Here’s a video post calibration, showing the tightness of the belt, then jogging the Z up 10MM and how it loosens up.
The calibration log was lost because the machine froze then overwrote the file when I saved again, but here’s the log of moving the machine in the video. Wondering if that “BL: -7.321” Is the issue. Maybe the BL calibration isn’t quite right? Maslow-serial 3.log (1.8 KB)
When it starts the bottom belts go fairly slack. After the first few calibration points they tighten up more. I can take a video tomorrow if that helps.
That sets how hard the belts pull during calibration and so it indirectly sets how tight the belts are when moving around since if we calibrate by pulling harder if there is any flex in the frame it will be taken into account.
Also have a theory that the magnets might be loose. I bought it assembled and one belt was broken, when I disassembled to fix it a magnet fell out of the plastic sensor gear. So I might pull it apart and re-glue them all.
I ran an extend and tighten cycle a few times and the axis were all close to zero, so that might not be it. However, would give some peace of mind.
Also, @bar do you have any spare top router clamp assemblies? The bolt heads popped through on the Z axis support side. Wonder if they need a little more material under the bolt heads in the next version.
Might have something useful here @bar — I replaced all 4 belts and recalibrated. It’s still loosening up the BL on any movement, but when I run the tighten command, it’s showing BL as about -8MM and change out. It happens after a jog and a re-tighten command, or hitting re-tighten multiple times in a row w/o a jog.
Post calibration log snippit:
Calibration complete
Calibration values:
Fitness: 0.5535789097949809
Maslow_tlX: -9.4
Maslow_tlY: 2408.7
Maslow_trX: 2930.5
Maslow_trY: 2406.4
Maslow_blX: 0.0
Maslow_blY: 0.0
Maslow_brX: 2927.1
Maslow_brY: 0.0
A command to save these values has been successfully sent for you. Please check for any error messages.
[MSG:INFO: Setting z-stop position]
[MSG:INFO: Calibration complete]
Jog: X100Y100
[MSG:INFO: Measured waypoint 0]
[MSG:INFO: Center point deviation: TL: 0.201 TR: 0.097 BL: -8.540 BR: -2.077]
[MSG:INFO: Center point deviation: TL: 0.201 TR: 0.097 BL: -8.540 BR: -2.077]
[MSG:INFO: Center point deviation within 15.000mm, your coordinate system is accurate]
Jog: Z10
[MSG:INFO: Measured waypoint 0]
[MSG:INFO: Center point deviation: TL: 0.174 TR: 0.124 BL: -8.363 BR: -0.252]
[MSG:INFO: Center point deviation: TL: 0.174 TR: 0.124 BL: -8.363 BR: -0.252]
[MSG:INFO: Center point deviation within 15.000mm, your coordinate system is accurate]
[MSG:INFO: Measured waypoint 0]
[MSG:INFO: Center point deviation: TL: 0.161 TR: 0.076 BL: -8.189 BR: -0.354]
[MSG:INFO: Center point deviation: TL: 0.161 TR: 0.076 BL: -8.189 BR: -0.354]
[MSG:INFO: Center point deviation within 15.000mm, your coordinate system is accurate]
That is a good indication that something in the calibration process which is going on with the bottom left belt, but it doesn’t narrow it down too much what exactly it is that is happening
It’s a second-hand unit, the belts and belt gear ring were pretty thrashed so I changed them last night. Inspected everything and made sure the gears were not too tight — seems to run better now. The frame is built to spec with some extra cross bracing and sitting at a 15° angle from vertical. I also updated the firmware to v0.83.
Do you think any settings could have persisted from the previous owner despite a firmware update? Looks like he had some trouble with it and possibly changed some settings. Here’s one of his threads M4 Calibration Questions - #47 by KDB
go through a few cycles of extending and retracting the belts. Look at the
values that are displayed for each belt after they are retracted. If any of them
are significantly non-zero (more than 1-2mm, usually they should be down in the
0.1mm or below range), then you have a problem with an encoder. It could be a
magnet not glued well enough, the magnet in the wrong place, or something else.
