VistaCNC P4-SE pendant stopped working

Yeah, I think I would agree with the 'agree to disagree' (pretty soon we can test how many times we can have agree in a sentence. ) . But let me ask this... Do you think it is a bug, then, that you can re-enable the Upper left (aka ESTOP, F1) button in the UI? Being literal, it seems like you are saying that it should be treated as a one-shot. Once you enable it, if you ever press it again, then you have to restart all of LinuxCNC.

My humble opinion is that this would be a poor UI, and I suspect that the community would agree or it would have been designed that way. The EStop button(s), regardless of where or how many you have, behave in the same way as that UI button.

For what it is worth, though... I am grappling a bit with the different states. As it stands now, EStop shuts off power to my servo amplifiers. To me, that is the absolute "you cannot move any axis or turn the spindle on". Machine On/Off merely enables the amps. I don't regard that as being nearly as "safe" as without power. I've toyed with the idea that the On/Off control power to the amps/VFD *and* enable the amps. In that mode, I'd have to decide what the ESTOP switch would do.

The mode of the original Hurco control was kind of like my approach. EStop powered off the amps/VFD, but that state was completely reversible by pressing ON. The machine had no pendant, although the discussion is more about EStop in general and not related to a pendant or lack thereof.

Whether you agree with my interpretation of ESTOP or not, the bricking of the pendant is not the manufacturer's desired behavior. I think that is perhaps the more important point here. Someplace, somewhere, there is a bug. That issue might be hardware, software, or both, but there is a definitely a bug.

JohnnyCNC: actually, I don't think the cable is shielded, at all... Or, if it is, the shielding definitely isn't carried through the RJ connector, which effectively means (in my mind) that the shielding isn't particularly beneficial.

Their wiring diagram of the EE_Box indicates that the RJ45 cable is shielded but I agree that what's the point if it is not carried through.
I'm going to see if I can make it carry through. I don't know if USB cables typically have shields or not. I would guess not.

LOL... So that is what it meant when it said "10' shielded cable", LOL. I skipped over the text and was looking at the connectors. That was partly because, well, one of my first thoughts was that the connector was damaged. I needed to remove the OEM rubber boot because I wanted a watertight RJ connector which required a hood that was incompatible with the rubber boot. Anyhow, the boot was glued on with some horrific glue. I considered that maybe I damaged with wires so I recrimped a new connector. In doing so, I accurately replicated the original, of course, and there was no shielding being carried through.

But, this is secondary to experience too. When I installed the new connector, I had to strip back the wire a bit. I wouldn't bet my life on it, but I really do *not* recall there being any shielding on the cable. I'm not sure I believe that bit of text in the linked image.

An excellent question, and something that Lee@VistaCNC asked this morning. The short answer is yes, within the stock OMRON part. But I need to validate that the diode is good. I missed that point.

However, I think there may be more at play here. One of the steps I tried a bit ago was to remove the COR relay (socketed, so very easy to do). If you look at the schematic clipping I sent, that would effectively render the START switch useless (I don't have a System ON indicator installed... part of the front console which is only a concept at present). Yet, pressing the START switch absolutely is correlated with bricking the pendant. I can't explain that bricking because back EMF would not be an issue. There is no inductive load (COR coil) being switched on in that case.

But, that does get us to a different point, also something I have been thinking about and that VistaCNC is suggesting)... An isolated 24VDC power supply. I'm presently using my computer power supply to provide voltages. A normal ATX power supply does not have 24VDC. I am using a buck boosting supply for that. I looked at the output with a scope and it looked good, but this whole problem kind of reeks of noise, so I'm going to dump the buck and spend a buck (OK, poor pun) to buy a separate 24VDC supply.

PS: I validated the wiring, although the terminal IDs in the clip I attached to an earlier post are incorrect. I had added an additional block that messed up the IDs... since corrected but not worth putting a new attachment for that.

ON EDIT:
Sorry, Andy, I missed your question about whether the behavior is EStop is the same regarless of where it is pressed (pendant or F1). I get the same behavior on both the mechanical switches, which is logical as they both are in the same circuit and subject to the same noise, etc. F1 on the screen behaves in an entirely different fashion and in the way I would expect and desire.

In my situation I have never noticed a correlation between pressing start and the pendant bricking but that could be I just never noticed it. I do have a separate 24v power supply for my field power and a standard ATX power supply running the PC, and a Non-regulated 48volt power supply for the stepper motors. All of which is in the same cabinet connected to the same 240volt supply circuit. I do feel that my issue is a noise issue. I can go for months with no issue and then it will flair up and drive me nuts.

I'm not sure we are talking apples and oranges here. My "START" switch is specific to my estop chain. Do you have a latching relay setup like the one I have in my schematic snippet that is attached to an earlier post from me?

As for this being a noise problem... I agree for my issues too. It seems more and more like that is the problem.

I'm sure your ee-box makes your cable becoming a nice antenna for emv. If I see correct it is nothing but plastics. And also if the USB cable would be a bit shielded, that is not a real screen. Try to get a good contact from pendant cable screen to your pc-case or the plug as short and thick as possible. Such screening can also minimize spikes from estop cables. You can also take two ceramic capacitors from each estop screw and connect them to pc case.
Maybe that is not the reason, but after that I'd say you excluded direct influence of emv to pendant as reason.
Mike