Hello Guest it is May 20, 2019, 01:04:20 PM

Author Topic: Phantom Limit Switch  (Read 1361 times)

0 Members and 1 Guest are viewing this topic.

Phantom Limit Switch
« on: March 17, 2016, 08:09:41 AM »
Breakout board is a parallel port board c10 is on it. No motion control, direct to PC.  We varied the debounce range from 1000-8000 because we could not successfully run the program more than two times in a row without the phantom limit switch. We then tried to isolate which limit switch that was creating the signal by turning off the pins in Mach3. However, it was sporadic on which one triggered it. The program runs when all of the limit switches are taken off. I also grounded all of the shielding from the input wires to try and eliminate any rogue signal from that.

We tested all of the motors and grounds for the machine which have continuity. We tried ceramic capacitors as per a suggestion which did not solve the issues. The CAP's are now removed. We are still experiencing a "limit switch triggered" almost half way through the program. We detached the proximity sensors from the machine and tried to run the program to see if vibrations were causing the limit switches to be triggered, it was not the case. The "limit switch triggered" message came up again during the process.

Machine is a Chinese KL-1212V with NEMA motors that we replaced. Purchased through Automation Technologies.  Wea re debating about swapping the VFD out with a machine that already works.

Please let us know if you have any ideas.

Offline Tweakie.CNC

  • *
  •  7,764 7,764
  • Super Kitty
    • View Profile
    • Tweakie.CNC
Re: Phantom Limit Switch
« Reply #1 on: March 17, 2016, 10:46:34 AM »
It’s impossible to be absolutely sure but from what you describe (having ruled out vibration) it is more than likely electrical noise from the VFD being picked up from the limit switches / inputs wiring (including e-stop, touch probe, etc.).

My opinion which is basically the same as others have said in your earlier thread…
If you haven’t done so already, create a single point Earth at the controller, connect this point to the incoming Earth and connect the shields of all inputs cables to this one point (leaving the other ends of the shields unconnected. Re-connect 0.1uF ceramic capacitors between each used input to the BOB and the single Earth point.
The wire from the VFD to the spindle motor also needs to be shielded and again the shield is connected to the single point Earth and left unconnected at the other end.
If found necessary, re-route the wire from the VFD to the spindle motor separately to the other cables. Check that you (or the Chinese) have not created any GND loops or Earth loops in any of the wiring (these make wonderful antennas).

Hope this helps,

« Last Edit: March 17, 2016, 10:50:00 AM by Tweakie.CNC »
Success consists of going from failure to failure without loss of enthusiasm.  Winston Churchill.
Re: Phantom Limit Switch
« Reply #2 on: April 20, 2016, 03:25:05 AM »
Had the same problem myself while machining aluminium GT2 pulleys.
It started when I switched from a single point cutter at 2400 RPM (40 Hz) to a 10-tooth cutter at 2700 RPM (450 Hz). The industrial microswitches used for the Limits started to misbehave, but were fine when the machine was stationary. They were also fine when machining plastic dummies.

I changed the spindle speed from 2700 RPM to 1800 RPM and the problem went away.

Micro-switches are specified for all sorts of conditions, but the one thing they are NOT specified to handle is vibration at audio frequencies. I believe some part of one of the switches was going into resonance and 'bouncing'. Changing the spindle speed meant the machine was no longer near that resonance.

Well, it worked very well for me.