Took a year, but finally finished assembling and testing my machine (2’x4’ / 2’x2’). I’m using the old style (not 525). Testing with a pen worked beautifully; perfect test patterns in high detail.
However, I’ve now connected a router and there is so much side-to-side movement in the middle assembly (2-3mm) that the pull from the router causes the cuts to warp in the direction opposite to the movement. I shortened the machine to 2’x2’ in the hopes that the problem was the 4’ length. No improvement.
On closer examination, the bearings on the middle are not flush with the bars, thus allowing torsional movement. There’s probably about 1/3mm of space between some of the bearings and the rod. This is enough space to cause 2-3mm of movement at the end of the router bit.
There does not appear to be a way to tighten the bearings against the bar. Is that correct?
Does the latest design (525) allow for tightening? Or somehow improve the rigidity?
Are any of the parts from the old model compatible with the new model (525)? Or would I need to reprint everything? It looks like the base (feet) are the same. Hopefully the x/y motor mounts are compatible?
Could be a few things here but lets start with the worst one. All the bearings should touch. If they don’t take it apart and put it back together and pay attention to when things go wrong. Each part is actually interference fit so with no tension on the tensioning bolts they should still touch. If that isn’t the case something is wrong. Meaning the xy parts alone should all touch and actually be a bit snug, as well as the XYZ parts individually.
Assemble the parts with the short bolts and bearings snugged up, put the tension bolts in with no nuts and keep them loose. Put in all 4 rails before you check for square or bearing contact or before tensioning.
Make sure your z axis is parallel. I’m not 100% clear with your description but it actually sounds like the Z axis is your problem, are you using the correct parts and are your rails parallel and not twisted?
Usually this happens when someone tries to square it without the Z axis in place or some other variation of squaring before complete assembly.
>>…are you using the correct parts and are your rails parallel and not twisted?
The z-axis lays flat on the table so I’m assuming it’s not twisted. The outside distance between the z-rails are 50.5mm on both the top an bottom so they appear to be parallel. The z-rails are a nice snug fit against the bearings. That said, I think you may be implying that if the z rails are not parallel then that could cause the x/y rails to be misaligned.
>>Usually this happens when someone tries to square it without the Z axis in place or some other variation of squaring before complete assembly.
I don’t recall actually “squaring” the middle assembly. Do you still have the instructions for doing this for the original machine?
Thank you for the excellent advice. I’ll do a re-assembly this week.
Cool, it the Z is in good shape you should be fine.
I don’t know what the exact dimension is for the Z width but if the bearings touch then you should be fine as well. I can double check but I have the old stuff archived.
Squaring the old and new is not very different, same bolts do the same thing. As long as you have used it a bit to break it in I think the second time around it should assemble better. I do things like swap a xy part or a xyz part but not both. Now that you know what you are looking for it should be fine. If you have more problems lets see some pictures.
So, I successfully, re-built, squared and tightened the middle. It seems that I hadn’t tightened the tension bolts at all (still loose from initial assembly). The x/y/x rails now fit snuggly against the bearings.
Unfortunately, this was a red-herring. The cut line is still warping as it moves along the X-axis.
This is what I’m trying to cut (using a 3mm cutting bit):
This is what get when I use a 0.5mm cut depth (just a light surface cut), hopefully demonstrating that there isn’t any major issues with the x/y:
This is what happens when I use a 3mm cut depth
This is 3 pass cut, 1mm per pass, going to a depth of 3mm:
As the router is cutting, it looks like the z-axis is flexing, or perhaps the router is moving. It’s very hard to tell, even watching it repeatedly on video. Note the use of the “universal adapter”… Perhaps it’s flexing…
The overall system is currently 2’x2’
Any suggestions? Can I offer any more information? Would the video of the cut be helpful?
20mm/s is pretty fast for a slotting move, try 8-12mm/s. Your Z axis speeds are too high try 3mm/s max on your machine is 8.4mm/s before the cpu crashes.
You can move up from there if your build can take it, you are using a rather large build and the old parts so start slow.
Thank you! I should have realized that the speed was a contributing factor. I’ll slow it down then work my way up to see how fast I can really go.
BTW, I didn’t really have the speeds at the levels indicated by the ESTLCAM screenshot. I’d overwritten the default max-speed values in the firmware. That said, my max speeds are still too fast; I’d set the values to facilitate pen movement, not router movement.
You can’t rely on the firmware to regulate your speeds. It tends to crash more than regulate speed. It is good with accelerations but not max speed.