VFD cutting issues

Long time reader, first time poster.
So I finally ran into a problem I can’t figure out. I ran a job yesterday engraving aluminum with a Amana 45771-K, 30 deg v with .005 tip. Everything worked fine except my tip was more like .02, so the letters blended together. Went to my local wood store and bought a new bit.

I loaded the bit into the spindle, mounted new stock, turned on the XXL, homed, set zero, loaded the gcode from the day before into UGS, turned on the VFD and hit the go button. It cut 5 of 17200 lines of gcode and stopped. Reset everything and it did the exact same thing, 4x.

After checking all the cables I set zero out in space, left the VFD off, and it went thru all the motions until I stopped it, 6-7 minutes. Set it all up again and hit go, 5 lines and stop. I’ve run the program 3x with the VFD off and all seems to be fine.

I’ve made no changes to the machine or the software in months. The VFD cable is shielded and grounded. Any thought would be greatly appreciated.

Thanks
Rob

The EMI imps are devilish creatures…

  • Did you go and re-check ground continuity on your setup (with a voltmeter)? Something may have come loose, a wire could have broken
  • Any chance you changed where the VFD/machine is plugged, or plugged in something new recently on the same line?
  • does it stop exactly at line 5 everytime you run the job with the VFD on, and it so can you post the first lines of the file ? Current surge upon starting the spindle ? You could try testing a much slower RPM (as an air job) and/or a longer RPM ramp up time (it’s configurable in the VFD params).
1 Like

Julien,
Thank for getting back to me.

I just checked the continuity between the spindle body and the ground lug in the VFD, between the shielded braid in the cable and the conduit, and the PWM and GRD wire from the board to the VFD and they are all good.

The VFD has been on the same plug since day one, shared with the band saw that hasn’t run at all today.

It was only the last time it stopped that I noticed that it showed line 5 of 17,200.
The ramp up time PD014 was set to 3 and is now set at 4

I’ll knock down the RPM and give it another go.

Here are the first few lines.
T1
G17
G20
G0Z0.8000
G0X0.0000Y0.0000S16000M3
G4 P5
G0X-0.0217Y0.0395Z0.2000
G1Z-0.0120F30.0
G1X-0.0231Y0.0381F50.0
G1X-0.0289Y0.0332
G1X-0.0332Y0.0309
G1X-0.0147Y0.0493
G1X-0.0128Y0.0528
G1X-0.0123Y0.0545
G1X-0.0383Y0.0286
G1X-0.0424Y0.0273
G1X-0.0112Y0.0585

Slowed it down to 14000 rpm and this is how far it went before it stopped.

I was mistaken it was 5 of 169,800

T1
G17
G20
G0Z0.8000
G0X0.0000Y0.0000S14000M3
G4P5
G0X-0.0217Y0.0395Z0.2000
ok
G1Z-0.0120F20.0
G1X-0.0231Y0.0381F50.0
ok
ok
G1X-0.0289Y0.0332
ok
ok
G1X-0.0332Y0.0309
G1X-0.0147Y0.0493

Weird…and I assume the spindle runs normally?
To narrow down the issue I would probably try and run the same job from another Gcode sender, from my experience they can sometimes react differently to communication glitches. CNCjs has been the most robust for me, I had a few events where it reported wrong commands, and yet allowed me to force-continue
Other than that, maybe trying to run with the VFD but with the spindle disconnected? (careful…)

Yes, the spindle runs fine. I just updated and ran CNCjs and got the same results. Runs fine with the VFD off and stops after a few lines with the VFD on.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.