GHOST? Noise? Bad Luck? Could use some ideas..

[Centroid_Tech]

Looking at your report, you have the Motor Revs/In set to 1 while the Encoder Counts/Rev is set to 8126 for the X and Y axes and 8216 for the Z axis. That doesn't seem correct. Please verify that as the steps for a stepper motor are typically a value of 200, 400, 800, 1600, 2400, etc. I also see that you have 0" set for backlash on all axes. Is this correct? I do see that spindle Nerd asked about backlash but I didn't see any replies from you stating that you verified the backlash.

[diycncscott]

You don't mention anything about what make and model drives you are using or how they are wired. External pulls ups? internal? none? Please post the make and model of the drives you are using, any specs/documentation you may have on them and a schematic depicting exactly how you have them wired.

[Gary Campbell]

That number, "8126" is suspiciously close to what is used for a 5mm pitch ballscrew on a drive that is set to 1/8 ms or 1600 steps per rev.

Try setting your steps per rev to 1600 and your revs/inch at 5.0801 {1 / 5mm (0.1969)} = 5.0801 revs per inch

My calcs return that combo as 8128 steps per inch, not 8126

[@dam]

Good catch on the Z steps. All three were supposed to be 8126. Oh the struggles of being a dyslexic operator

The 8126 value comes from how the machine was configured before the Acorn conversion. With that value it is spot on with the dial indicator.

The drivers are set to 1,600 steps/rev. I don't know what the ballscrew is, so that's why I capture the steps/inch with a multiplier of one. Same difference right?

This machine has never had backlash comp configured and there isn't detectable backlash with our dial indicator resolution (.001).

Looking at your report, you have the Motor Revs/In set to 1 while the Encoder Counts/Rev is set to 8126 for the X and Y axes and 8216 for the Z axis. That doesn't seem correct. Please verify that as the steps for a stepper motor are typically a value of 200, 400, 800, 1600, 2400, etc. I also see that you have 0" set for backlash on all axes. Is this correct? I do see that spindle Nerd asked about backlash but I didn't see any replies from you stating that you verified the backlash.

[@dam]

Thanks Gary. I think you might be onto something here. I'm going to the shop in about 5 minutes and will give this a try. Thanks!

That number, "8126" is suspiciously close to what is used for a 5mm pitch ballscrew on a drive that is set to 1/8 ms or 1600 steps per rev.

Try setting your steps per rev to 1600 and your revs/inch at 5.0801 {1 / 5mm (0.1969)} = 5.0801 revs per inch

My calcs return that combo as 8128 steps per inch, not 8126

[diycncscott]

You should always set your steps/rev to what the drive is configured for -in your case 1600 steps/rev. That is a finite number that is determined by the capabilities of your drive. Does it "all come out in the wash"? Maybe, maybe not. Again, this is limited by the capabilities of your drives and you could quite easily find your system in a state where your max rates, step/rev and/or step frequency exceed what your motors/drives can reliably respond to.


Rev/unit = 1 Steps/rev= 8126 = 8126 steps/unit

=

Revs/unit 5.07875 (8126/1600) Steps/rev = 1600 5.07875*1600 = 8126 steps/unit

[@dam]

Well unfortunately I got the setting changed to this and still was off by .04. Thanks for this info by the way Gary! I was able to confirm it was a 5mm pitch ballscrew. We had just always used the parameters the machine came with because it "worked"

Thanks Gary. I think you might be onto something here. I'm going to the shop in about 5 minutes and will give this a try. Thanks!

That number, "8126" is suspiciously close to what is used for a 5mm pitch ballscrew on a drive that is set to 1/8 ms or 1600 steps per rev.

Try setting your steps per rev to 1600 and your revs/inch at 5.0801 {1 / 5mm (0.1969)} = 5.0801 revs per inch

My calcs return that combo as 8128 steps per inch, not 8126

[@dam]

Drivers are Leadshine DMA860H (manual attached). They are wired based on the attached Acorn Leadshine wiring schematic (attached). I have not made a new wiring diagram to depict our exact machine.

I believe next steps will be swapping the driver and motor from the Y axis to see if I get the same behavior.

Thanks!

You don't mention anything about what make and model drives you are using or how they are wired. External pulls ups? internal? none? Please post the make and model of the drives you are using, any specs/documentation you may have on them and a schematic depicting exactly how you have them wired.

[Gary Campbell]

adam...
I had better luck using the DB25 connections than the open collector terminals.

[@dam]

Still no luck. Have tried switching driver and motor with another axis. Checked and rechecked all wiring. Broke down the Z axis to looking for any binding.

Also spent the night temporarily wiring up the old Mach4 BOB and smoothstepper. Ran the same job 3 times and didn't lose a step.

Have now run the test on the Acorn 26 times. Only new development is the amount of lost steps is less with smoothing is set to precision router instead of contour router. Does anyone know what smoothing settings would be equivalent to the Mach4 constant velocity mode?

Thanks-

Adam