Retrofit My Build Looking at Switching from Mach4 to Centroid System

[cnckeith]

are you using the Teknic RAS feature in there setup software for the axis motors?

Sorry I didn't see this post.

Yes the engineer that logged into my servos remotely set RAS up.

RAS introduces a lag that shows itself with Laser applications (XYZ)
when using a diode laser you can compensate for the lag introduced by RAS

with this added feature in version 5.24+ PWM Laser Delay Fine Tuning Parameter 424 added. Useful for compensating
for millisecond delays introduced when using RAS with Clearpath servos.

page 14
https://www.centroidcnc.com/dealersuppo ... ds/326.pdf

[Onawa220]

I haven't updated to the latest version of CNC12 since we did the custom PWM files. I don't suppose the update incorporated those into it?

Also I ran the X & Y tests and they looked pretty good to me but I could be wrong.

[Onawa220]

Use the GRBL output and then "Save to G-code", along with the custom start/end code option in machine settings. Then if you need some further editing of the saved G-code, open it in Notepad++ and make a macro to do any of that. My macro starts at the first line and automatically makes all the needed changes and then saves the file.

Further info...viewtopic.php?p=72920#p72920

I am still working through getting everything set up but I ran a test on a file I created in Lightburn then did the manual editing myself since I haven't taken the time to figure out how to run the macro yet. The laser wasn't turning off as it moved to other areas so I'm missing something. Maybe the macro would add or remove what needs to be done to turn it off when moving and not supposed to be firing?

This is a link to the gcode I used. https://www.dropbox.com/scl/fi/fyr89h3g ... 4hgxk&dl=0

[cnckeith]

Diode Laser setup TB was update a few weeks ago, page 14 has the section with the new clearpath specific feature
https://www.centroidcnc.com/dealersuppo ... ds/326.pdf

[Onawa220]

Diode Laser setup TB was update a few weeks ago, page 14 has the section with the new clearpath specific feature
https://www.centroidcnc.com/dealersuppo ... ds/326.pdf

Sorry, what I was asking about is if I update to the latest version of CNC12 will I need to overwrite the files with these files to turn off the spindle etc when using the laser or did this update incorporate the changes in the files linked?

download/file.php?id=30982

[Onawa220]

That did it! The laser no longer goes to 100% at the end of the cut file!

The only "issue" now is it is still controlling the VFD at the same time. That seems strange. Is it some kind of cross talk between the spindle and PWM connections on the board?

Can try the attached PLC, that disables Analog1 when M37 (Laser) is active. This should prevent analog voltage being sent to the VFD during laser operations.

But in general, the reason it was commanding the analog is they both share the Same S command between the Analog and PWM signals. So S commands for the laser also change the analog for the spindle. Analog1/PWM1 are linked and so is Analog2/PWM2, i can disable Analog1/PWM1 during M37 through the PLC, but would also mean the PWM1 is also disabled. Though shouldn't be an issue as its recommended to connect the laser to PWM2 anyways.

Allin1Chris are these fixes incorporated into the newest version of CNC12? If not is it safe to use these files again to overwrite the stock ones?

[Allin1Chris]

Allin1Chris are these fixes incorporated into the newest version of CNC12? If not is it safe to use these files again to overwrite the stock ones?

Please Try V5.30, It should contain the fix for this.

[Onawa220]

Allin1Chris are these fixes incorporated into the newest version of CNC12? If not is it safe to use these files again to overwrite the stock ones?

Please Try V5.30, It should contain the fix for this.

So far so good. Thank you so much!

[Onawa220]

Allin1Chris are these fixes incorporated into the newest version of CNC12? If not is it safe to use these files again to overwrite the stock ones?

Please Try V5.30, It should contain the fix for this.

I have run into an interesting little bug in 5.3

In version 5.2 I did not have any lash compensation set up and doing the zig zag X & Y tests for my laser things were almost perfectly lined up. (see next two pics)

Today as I’m getting things setup and fine tuned in 5.3 I did the same test and they are now way off AND I see that in the Y test the laser stays on while moving. In the X test it does not.

I went back and looked at my first zig zag test from 5.2 and I can barely see that it was also firing in this test while moving but barely and it’s hard to see in the picture.

I’ll follow the guide on how to how to adjust for my Clearpath servos with RAS enabled but it seems strange that I had no compensation set up in 5.2 and it was lined up.

I’m also attaching the last report I ran yesterday before switching to 5.3 and I’ll include the report I just ran in 5.3 with all the same settings.

report_0008DC111213-0116240167_2025-03-28_09-44-46.zip

Last 5.2 Report

(1.12 MiB) Downloaded 1 time

report_0008DC111213-0116240167_2025-03-29_08-12-58.zip

First 5.3 Report

(1.04 MiB) Downloaded 2 times

ZigZagLaserSyncTest-Y_Axis.cnc

Y Test from 5.3

(599 Bytes) Downloaded 2 times

ZigZagLaserSyncTest-Y_Axis.cnc

Y Test from 5.2

(599 Bytes) Downloaded 3 times

[Onawa220]

To add to the issue.

My lash settings in CNC12 were all 0.
I checked my backlash again and had almost none as I expected since I tested it when I initially switched to my AcornSix. It measured at .00005 on X and Y. Not even enough to make any changes in the lash settings.

In P424 If I went up to 12ms. The X lines were pretty close, I could have fine tuned it more, but the Y was way off and I don’t think they really got any closer together as I went from 0 up to 12ms.

I then put the P424 back to 0 and went in and changed the lash settings until I had it the X & Y lines pretty close to perfect with 0.025 on the X and 0.045 on the Y

I then tested my backlash again and predictably it was way off.

So it’s almost as if P424 only changes the X and not the Y.

And to prove to myself that I wasn't imagining things I switched back to 5.2 and re-ran the X & Y tests.

I confirmed no lash compensation setup. The interesting thing is the Y doing long lines this time but not the last time I ran it in 5.2