Author Topic: Random E-stops Doesnt Seem to be Noise Related?? Fix it and I'll pay you a BUCK!  (Read 4498 times)

0 Members and 1 Guest are viewing this topic.

Offline Wilde Racing

  • Active Member
  • Posts: 58
    • View Profile
Hi

I have a old Kondia CNC that had a Dynapath S10. I have recently retrofitted with Camsoft and Galil DMC-1842 and ICM-2900. You can see my other thread in this section. I have slowly been making progress with the conversion to Mach3. ( I never had camsoft running, I didn't even try.) I could not figure out the spindle control so I got a cheap Haunyang and tried to run one of the mach plug-ins. I could not get the spindle to run more that a few seconds before shutting down with a false E-stop. I tried changing the Debounce and it seemed to have no effect. I replaced the spindle harness with a shielded VFD style harness. The harness changes seemed to have no effect. I finally gave up on the plug-in and the wired the spindle threw Galil. Now I have had it working for a couple days with no false E-stops. then all of the sudden I started having issues again... :( Changing the Debounce from 10 to 100 to 1500 to 2500 none changes the issue. The E-Stop only happens when the spindle is on and I'm running a program. I have been dry running ( no cutting) with some wizards to test settings.

A few things I've tried.

Moving the computer away from the spindle ~ Seemed to help at first
Placing a piece of sheet metal between Computer and Spindle ~ Seemed to help at first
Debounce settings ~ No change
Rewired VFD with VFD shielded wire ~ No change
Manually controlled spindle with Galil and relays ~ Seemed to help at first.


Any other thoughts or ideas?

How close is too close for the computer to be to the spindle? The VFD is n the other side of the mill but the computer is fairly close to the spindle motor.

Offline Wilde Racing

  • Active Member
  • Posts: 58
    • View Profile
Make it ten bucks!!!! I've spent every free moment I had on this thing and I'm so close to being able to use it!!!

Offline Tweakie.CNC

  • Global Moderator
  • *
  • Posts: 7,641
  • Super Kitty
    • View Profile
    • Tweakie.CNC
Unfortunately there is no one answer that suits all situations – electrical noise (and I am sure that is the problem) is bit of a chameleon.
A couple of suggestions…

1) A single point (star) EARTH with all cable shields taken to this one point (shields left unconnected at their far ends).
2) RFI filters fitted to both the VFD supply input and the machine controller supply input.
3) 0.1uF capacitor fitted between each used breakout board switch Input and GND.
4) Routing input signal cables (switches, etc.) away from ac power cables.

I don’t suppose 10 bucks buys much Beer these days but that always seems to help me with problem solving.  ;)

Tweakie.
Success consists of going from failure to failure without loss of enthusiasm.  Winston Churchill.

Offline RICH

  • Global Moderator
  • *
  • Posts: 7,300
    • View Profile
Quote
How close is too close for the computer to be to the spindle? The VFD is on the other side of the mill but the computer is fairly close to the spindle motor.
The answer is rather simple.......but elusive ::)

Far enough away such that any EMI / RF field is not strong enough to affect
the recieving electronic device. But, that statement assumes that it is only the field that is causing a problem and nothing was done to control / minimize / reduce the EMI.

Try "sniffing" around to get a flavor of intensity / frequency.

RICH

Offline Wilde Racing

  • Active Member
  • Posts: 58
    • View Profile
Thank you both for the suggestion!!

What I have figured out is that my brain is the issue! Particularly the brain that controls my three external buttons. On is a double pole estop button and the other two are momentary switches that are wired to cycle start and feed hold. All three ground input pins 1, 2, and 3. I also have my estop wired to shut down the power to the amps via a relay. When I disable this brain I get no issues. Enable it and I get my false estop issue.

How do I fix it? Installing a cap on the estop input as Tweekie suggested? All three switches are in the same shielded harness with a single drain wire properly grounded. Why don't I get and false feed holds??

Thanks guys!!!

Offline Wilde Racing

  • Active Member
  • Posts: 58
    • View Profile
Just to close this thread out and to help other with similar issues, the cause of my issuse was the 100 pin connector.  It still causes some problems and is very finicky.  I have both ends ziptied down to restrict movement and the system have been performing just fine. The cable and connectors look fine, i went as far as removing both ends and checking individual wires....

Offline Tweakie.CNC

  • Global Moderator
  • *
  • Posts: 7,641
  • Super Kitty
    • View Profile
    • Tweakie.CNC
I am pleased that you found a solution and thanks for returning to the thread and posting details of the cause.

Tweakie.
Success consists of going from failure to failure without loss of enthusiasm.  Winston Churchill.

Offline bbutcher

  • Active Member
  • Posts: 51
    • View Profile
I made a cable clamp out of aluminum angle that I attach to the computer and to the controller to hold the connector solid.