I’m being particular with this setup. Using M666, I can square the machine in the XY plane. However, the machine does not sit flat on the rails when it is square. Looking down the rail, only one bearing on each wheel roller sits on the rail.
I’ve determined that the gantry is not square to the roller blocks. This is difficult to measure, but this is the logical issue through deduction and how I solved the issue. The flat material is MDF and aluminum, so the culprit isn’t likely that. Several places could introduce skew.
Ultimately, I shimmed one of the roller blocks away from the XZ plate, the thickness of 2 zip ties. It is dead on square when it naturally squares itself in with the steppers disabled. And homing to the end-stops (with some 666 offset because the stops aren’t perfect), it’s square enough that I can trace an L-square perfectly.
I’m thinking of a better overall solution to adjusting any mechanical inaccuracies. For shimming, the two shim points would be between the rail rollers and XZ plate or between the gantry end clamp and the YZ plate. My square issue is likely between the end clamp and the YZ plate. This could have been a printing issue where the edge lifted from the print plate. This is also the best place to make an adjustment that doesn’t throw anything else unnecessarily out of alignment. Ideally, a 3d printed shim to maintain surface contact.
It’s difficult to measure with everything in place, but one side was a bit shorter, probably the distance I would shim. The short side correlates with my skew direction and what the current shims would correct. Everything seems to point to that end gantry clamp not being square on the side that touches the YZ plate.
No shims should be needed. Have a good look to see if there is anything that needs to be corrected. I would venture a guess that your rails might be hitting your metal plate? Make absolutely sure they are not touching, this is mentioned in the instructions.
That would most likely kick the toe out. You did not way what dim was off though.
The mis-alignment is in the Y dimension. So on the rail side of the machine, as the router moves right to left (away from the rail), the router moves further away.
|
|\
The rails are not hitting the plates. I made sure to adjust the strut length to keep a slight gap.
I put a straight edge against the aluminum plate and the gantry is not square to it. That connection point is the gantry end clamp. Everything looks tight to each other.
My other thought is the strut plates. I calibrated the machine really well before cutting them in length and square. By design, it should hold the gantry rock solid square, however, if the strut plates have any skew in the milling, and if one of the plates were flipped end to end, that could throw it out of alignment.
I’m going to tear it down again and see what’s going on.
I tore it down again, and it’s measurably not square at the plate to gantry connection.
I’m re-printing the end clamp upside down. This will put the top of the print against the aluminum plate. The top of the 3d print will be the most square, so any hot-plate lifting issues or whatever will be nullified.
I will also screw down the strut after squaring it to itself. That’s really the point that determines what square will be. There’s some wiggle room there until it’s all tightened down. We should probably add some guidance to check the physical square before tightening this point. I probably could have simply loosened the strut, tapped it back to square, and tightened it up again.
I re-printed the end clamp, and put everything back together. It was still slightly out of alignment. I was able to loosen up the strut screws, push the clamp square, and tighten it back up. This held.