Closed loop stepper driver configuration help

• (free) Alignment
  • Check axis alignment and free movement.  You haven't said what this machine is, or whether this is a dovetail, box, or rolling element way machine.
  • Misalignment in the axis way components or improperly adjusted gibs can cause positioning inaccuracy - either in certain locations or across the whole length of travel.
  • Check ballscrew alignment.  Is the torque required to move the axis the same down the whole length of travel?  Does it bind up anywhere or at the ends?

Assuming you can't fix it by just tightening mounting screws or similar free things, you can throw money at it and replace any component or assembly that has backlash or is contributing to the inaccuracy.

• ($) Bearings
  • Buy some angular contact bearings and replace the (almost certainly) deep-groove ball bearings in the fixed end ballscrew support.  That won't be a ton of money - you don't need to buy top-of-the-line AC bearings... anything that isn't crunchy will be better than DGBB's.
• ($-$$) Ballnut
  • Re-ball the ball-nut with oversized balls (not expensive, but tedious and not for the faint of heart)
  • Replace the ballnut and hope the new one has a bit of preload
  • Replace the ballnut with a double-nut that has some preload (assuming you don't mind possibly losing a bit of travel)
• ($$$) Ballscrew assembly
  • Buy a new or used ballscrew assembly of higher quality
• ($) Stretch Goal
  • Buy a better quality coupler that has less rotational compliance or backlash

Not that it's possible to determine much from these but here's what they look like, on the Y axis. 

It's a LyCNC cast Iron 4020 mill, I'm piecing together information as I find it but it's not been easy.

X=380MM,Y=220MM,Z=280MM

thanks again

Those are significantly nicer components than I was imagining.

The linear rails appear to be properly side-clamped against ground registration surfaces with those screws & dowels, and I believe the rails are name-brand Hiwins.  I'd be surprised if there's any serious misalignment-induced binding.

You may actually have AC bearings in the combination ballscrew/motor mount.  And your coupler is a zero-backlash type.

I'd remove the motor and proceed with checking screw fixed bearing slop.  If there is any you can try snugging up the nut.  The machine looks clean but used, so it's possible the bearings are a little worn out.  On heavy or fast machines ballscrew fixed bearings take a beating.

After that, the nut, nut adapter, and motor mount screws are now (in my mind) the most likely culprits if there's a mechanical problem.  Likely the nut & balls, TBH.

Aliexpress has a few vendors of Taiwanese "C3" double-nut ball-screws around $400-$800 depending on length & dia. I don't know what the machine is, but it looks nice enough under the hood to be worth a set of ground ballscrews if you can scrape up the money. 

I conducted a few more test cuts on aluminum, and it's not ±0.10 mm; it's consistently +0.20 mm. I'm not sure if this indicates that an adjustment can be made on the software side or not?Apart from always being +0.20 mm, the variance is about ±0.04 mm.

I can't imagine the whole thing is flexing that much under cutting pressure, but I'd say next step is to mount a DTI on the table and touch the stylus to the spindle nose.

Then shove on the head and see if there's deflection.  Repeat for Y in both directions.

How much of a cut are you taking to get that much error?  Are you measuring the error by comparing the part vs. programmed size?

I'll try that when I can get a hold of a DTI. For now I went redid the gcode with a re-done tool table, and it seems to make the biggest difference so far. I'm now getting about +- .05 mm

I still think it can do better, but for now I don't want to take it apart it.

I probably need to test the spindle run-out as well at some point.

thanks for the advice it's been helpful.