Random E-Stop with CNC12 / Avid CNC with stepper motors

[corbin]

In your picture of the Alt-i diagnostic window, note that OUT7 (your "NoFaultOut" output) is red (off), even though INP1 (your "EStopOk" input) is green (closed, okay). That means there is some other fault condition still present. You may need some help from Avid or Centroid to pinpoint what fault that is, if there is no accompanying fault message to explain it.

Ah, so the NoFaultOut should be green....I guess my question is what would cause that to happen. What can I look at to find that out? I'll poke around in the code and see if I can figure anything out myself.

[corbin]

Is your estop switch/button of good quality or is it an Amazon product? Is it set up as NO or NC?

It's an Avid CNC, which has pretty good quality products. It's NC and opens when the e-stop is activated. If it was the switch...wouldn't flipping it reset it?

I was trying to rule out a defective switch.

As Marc pointed out you have other issues not yet defined. You have pressure low errors all through your error report along with tool touch off trips between the low pressure alarms. I'd spend some time trouble shooting those inputs.

That said your steps/rev are outside the sweet spot for a router. Do you have any gear reduction on this machine? This is another problem for another day. Here's some reading for you until then:

viewtopic.php?p=75374&hilit=sweet+spot#p75374

Yeah, I was suspect of the switch, but it doesn't seem to be the issue.

The low pressure errors are warnings; I turn the air off at night, and it gives me the warning. My ATC tool change code/macros will not do a change if the air is not on, but everything else works fine with that warning.

The Tool Touch Off Trips are another warning; Avid's touch plate is sensitive and trips when the machine is jogging. But it is just a warning, and not an issue. (they are also aware of this problem).

The steps/rev: This is a standard Avid CNC with stepper motors, and comes with their pre-determined ratio. I'm surprised that it is outside of the "sweet spot", and I would like to tweak it, but it isn't something I can easily do without a lot of re-work. This was discussed a bit before when I brought up some questions about smoothing: viewtopic.php?t=10920&start=20

What other inputs/outputs can I look into to try to solve this problem?

I appreciate the help and ideas!!

Thanks,
Corbin

[corbin]

...and if you look through my log, you'll see spindle errors too. Those happen if I forget to turn on my spindle box before my control box. That, and one time I did stall the spindle..

[cnckeith]

figure out how the estop switch is wired/routed and trace out all connections to find a weak one.

[Gary Campbell]

...and if you look through my log, you'll see spindle errors too. Those happen if I forget to turn on my spindle box before my control box. That, and one time I did stall the spindle..

The AVID system has virtually eliminated a number of sensors/signals that I feel should be included on ATC systems for proper use. This "dumbing down" of the control system to make it one size fits all for router, plasma, laser, ATC, etc. by combining any and all faults into one signal may be appreciated by inexperienced newbies, but surely has to be looked at with a great deal of skepticism by those in the know.

[corbin]

figure out how the estop switch is wired/routed and trace out all connections to find a weak one.

Yeah! It is easy to trace the wiring; I'll double check it. But if it was this, I would have seen it on the PLC Diagnostic Screen, right? And Disabling it would have let me reset the machine in software without a reboot of the Acorn controller (right)? But I couldn't get it to reset..

[corbin]

So, what can cause OUT7 (Setup to "NoFaultOut" output) to be Red (IE: an error is present)?

[corbin]

So, what can cause OUT7 (Setup to "NoFaultOut" output) to be Red (IE: an error is present)?

I'm guessing I need to use PLC Detective the next time the fault state is happening and look at all these things that set SV_STOP:

IF !EStopOk_M || PLCFault_M || SV_STALL_ERROR || SpindleFault_M || LubeFault_M ||
AxisFault_M || ProbeFault_M || OtherFault_M || DoorShutDownStop_M || THCError_M
THEN SET SV_STOP

Unless there is a way of knowing that after that via some log somewhere.

[cnckeith]

yep!
estop condition can be caused by a wide variety of items as you can see.
99% of the time its a loose/cracked/broken wire/switch/terminal/landing

[cncsnw]

I'm guessing I need to use PLC Detective the next time the fault state is happening and look at all these things that set SV_STOP:

IF !EStopOk_M || PLCFault_M || SV_STALL_ERROR || SpindleFault_M || LubeFault_M ||
AxisFault_M || ProbeFault_M || OtherFault_M || DoorShutDownStop_M || THCError_M
THEN SET SV_STOP

Yes, exactly.

If the problem (a fault condition that cannot be reset with emergency stop) is happening while you are at the machine, then you can use the source code view to look at the above logic line, to see what condition is causing SV_STOP to remain set.

If the problem is not happening at the moment, but you want to see what fault conditions occur, in what order, when a problem does happen, then you can use the logic analyzer in PLC Detective. Set it up to watch all of the various fault bits, and maybe trigger on a rising edge of Stop_M (a memory bit which mirrors SV_STOP).

For some general guidance on setting up the PLC Detective Logic Analyzer, see http://www.cncsnw.com/PLCDetectiveHowTo.htm