I have built the LR4 a few moths ago and this is my first CNC machine so I might be missing some basic concepts.
I use a draknife to cut a 1.2 mm plastic sheet. I use a 3D printed vacuum table that I put on top of my spoilboard. I have surfaced both the spoilboard and the vacuum table. I am still facing precision issues on the z axis with differences up to 0.5 mm. This is more pronounced on the x axis than the y axis. Do you thing I should adjust the z axis by tweaking the homing distance? Isn’t the surfacing supposed to take care of this issue?
all rails look pretty clean. I am working on a 2000x150 piece and the difference is pretty consistent with x max direction being the lower as well as being more pronounced towards the center of the y axis then the extremities.
I’d start troubleshooting this by bistecting problems. First question is to compare the surfaces of the table and spoil board to be sure they’re actually flat. For me, when I do this I get a metal straight edge (I like to use a metal level for this). I sweep the straight edge over the table surfaces with a light behind it and look for gaps to show me whether or not it’s truly level (planar, really as it’s a 2D surface)
There’s a bunch of ways that you could have surfaced something and it ends up not fully planar.
Let’s get that data then we’ll talk about using your system to probe / map the Z to understand what’s out of whack.
I probed at a regular distance along the Y axis both on the x min and x max. The highest probing gives -73.740 and is located on 0,0 the lowest probing is located in the middle and on X max resulting in -75.070. I therefore have a difference of 1.33 mm which when cutting the plastic sheet which is 1.2 mm thick is of course an issue. Overall x min side is higher than x max side. On the y axis it starts high, goes down and then back up (on both X min and x mix sides).
I have not surfaced the base yet. I will probably do it by removing 1.5 mm and zeroing on the highest point.
At y=0 x min is -73.740 and x max 73.905. I did here the manual z leveling by adjusting the stop switches. and 0.165 mm is quite good (this is also the min difference between depths on the same y)
The biggest difference I have between depth on the same y is 0.68.
So should I correct on the software side Z max by 0.165 or more?
My table is built on 6 2000 mm 2020 profiles. 2 of them under the x min, 2 under the x max and 2 under the cutting area. They are connected between them by smaller 2020 profiles with L brackets. The MDF if attached to the 2020 profile with tee nuts. The whole structure is sitting on a bench.
Maybe the bench is not 100% flat and the 2000 mm 2020 profiles are flexing in the middle. I could try to add 3 2020 profile along the width one in the middle and two towards the extremities.
I ended up adding 7 extra 2020 across the x axis along the y axis. It solved the issue. I measured every 300 mm along the y axis on both x min and x max. Now the max difference between two points on the same y is 0.3 mm and 0.5 the absolute difference. I can probably adjust 0.1 or 0.2 via the software and live with the rest.
I fully rebuilt he table as you can see from these pictures:
