@Brandon_Satterfield I’ve put a half dozen screws in the sheet around the bed, seems pretty strongly held in place. If I try and shift it my whole bench moves.
Since I’m air cutting at 7mm to start and then cutting down to zero do you think the bow in the alu would be a problem? I can’t probe because thats just not working on TinyG G2 for me.
I’ll be putting a pair of 7001C angular contact bearings in it. This piece is actually just a test to see if my sizings are right - I was probably going to make a stack of three of these to hold two of the bearings stacked up, with a lip on the last one to keep it all in place. The real pieces I want to make will be cut from a single 1" piece of alu I expect. This is all for my somewhat insane Burning Man project this year, a triplet of 8ft diameter double pendulums with hundreds of LEDs inside 
@Brandon_Satterfield that cut sounds much better than mine too, what hz did you have the dial at, or is this one of the computer controlled ones?
Computer controlled but it should be at the 250 range.
Btw bud I send you the file I ran. What you saw was exactly dropped in, no modifications.
Ok cool, I might just give that file a run, even without the probing it shouldn’t be too far off.
I also just went around the machine again checking for any wobble that could cause this. Discovered the right Y axis had almost 1mm of play in it, checked the setscrew on the backlash compensation nut and found that it was missing its locking nut. Tightened it up a bit and shoved a nut on there, hopefully thats what was causing all of these weirdnesses. Going to do another run now, wish me luck!
Good luck brother. As noted that machine I ran it on is the oldest now and loose everywhere. Well, probably not a mm anywhere. :-).
I don’t autolevel. When you do, what is your Delta Z, that’ll be a good indicator if your dealing with some rough metal.
When I managed to get a probe pass to work (one in like 10 actually succeeds) I was seeing a variation of about 0.04 to 0.07mm. Not a lot really.
Once again the metal defies me. I’m still getting a lot more chatter than you have in your video @Brandon_Satterfield despite playing with speeds up and down. There’s gotta be something else wrong here, I checked the runout on the bit and its tiny now 0.0005" if that. I guess I could have some misalignment on YZ i suppose, I’ll go check that again.
Dang dude. That’s running the same file I sent you?
Methodical approach. Based on where the chatter is at, which direction is there play? X or Y?
Well I had thought that the chatter was just because of the tab, but in that shot you can see its chattering in top right and in bottom left of the circle. So that would imply that its both axes?
I did that was at 0.6mm DOC, with all the same settings as your file except I did Z0 at bed, and stock surface at 7mm so that I air cut the first little bit or so to avoid a deep plunge into the bowed surface. Maybe I’ll try it with Z0 at work surface next time.
So I stuck a bit of z shaped tool into the thing, I’m not sure how good this collet is so I don’t know how accurate this test is.
But this seems to indicate that I have misalignment somewhere, but its strange that it is essentially the points where I’m having chatter just flipped around the Y axis.
Thinking on this a bit more this seems to indicate that I have a tilt going from +Y +X towards -Y -X, i.e. its tilting forwards around Z (expected due to spindle weight) and also tilting a little towards to the left around Y. This would I think explain the chatter in the top right and bottom left of the part. When the cutter breaks through Z0 in the top right the tooth will be digging deeper into the wall at that bottom edge because of this tilt. I guess I need to shim my axes…
Alright. There is the issue.
I’ve been concerned since the beginning about the multi-cut pieces with extrusions embedded. Dont shim the machine yet, shim the work piece.
Haha, too late, I’ve spent the last hour adjusting all the wheels on the gantry. I’ve managed to reduce the error front to back to 0.5mm over 172mm (the diameter of my test probe thingie). That is a misalignment of 0.175degrees on the Z axis around X.
I’ve got a few more wheels to gently tighten back up now without ruining my alignment but after that I think I will give it another run and see if that helps…
If it still doesn’t work I’m not sure what I’ll do.
The spindle weight should be countered. This is the reason your spindle clamps are sucked up to the Z as far as possible. A lot of work in that area.
Sweet, if it’s done it’s done. Would love to see if in action.
Using the file I sent?
I will do I think yeah, will zero the top of the work piece and just use that one direct.
Once again, failure.
I did it your way this time @Brandon_Satterfield though I did regenerate the gcode since I wanted to rotate the tabs a little to prove it wasn’t the tabs at fault, and I think its not given the result here. It still chattered in the same places and this time ate one of the tabs which was now in the path of the chatter.
Something else went wrong this time though and I think I messed up the gcode…as it was doing a long retract and plunge at each tab. Can you send me the cambam file with the modifications you made so I can take it from source?
