new OB XYZ Probe issues

Ok, so after the machine rebuild I just did, I'm now having issues with my probe.

Here's what I did:
Table and all mounting changed from wood to v-slot
Added relay chain (bb coolant output to 24v coil relay activating 110vac to control a contactor capable of running up to 13 cont. amps, yes 12awg copper wire) to be able to activate 110vac to start my dust collection system using gcode
Changed from a very DIY y axis track system (2" EMT as rails with 3d printed *plastic* carriages with bearings rolling on the EMT... had a lot of flex) to a more robust, reliable, and most importantly rigid HGR20 linear rail system.
Upgraded y axis belt tensioners
Managed external cables
Mounted my "special function" buttons pendant to the frame (reset button for VFD; kills power to contactor coil which activates mains to VFD. Relay is a timed relay set to about 25 seconds to allow the VFD to power off fully)
Removed unused external stepper drivers and associated components from inside panel (no longer needed, V3 used dual Z steppers. Might change back to that one day, not sure.)
Wired relay 2 to control direction of spindle (have disconnected and restored to original wiring configuration)

Anyway, when I got set up to do a first probing sequence to make sure I had enough clearance between where the spoil board and the from rail, and I noticed that as soon as I put the magnet on the collet nut, the light came on... I had this issue before, but it worked with the spindle disconnected, and eventually I changed something (not sure what...) and it started working properly.

After reviewing several threads here, I have come to the conclusion that I MUST have GND and earth ground connected somewhere.

I have disconnected every limit switch, so I don't believe its a limit switch isolation problem. I have also disconnected every motor including the spindle. No change.

I have double checked that there are NO GNDs going to earth ground, and no earth grounds going to GNDs.

With the exception of possibly a nick in a DC cable with a GND signal, I cannot think of anywhere else the two could be interacting. (I haven't noticed one, just hypothetically)

Anyone have any other ideas? The probe did work fine before.

Just to give you an idea of my setup, the enclosure is a metal enclosure with a interior mounting plate. I have mains (220v split phase from dedicated 30A breaker in panel) coming in through terminal block, which splits there and routes to various components, with both phases going to contactor which controls the VFD. Earth grounds are all to a ground bus mounted the mounting plate in panel. ground from VFD (Not spindle) also is wrapped around a couple of large ferrite rings to help reduce EMI, especially to the rest of the house. Most wires/cables have a ferrite choke.

So for the blackbox... Is any part of the case attached to GND? I don't think there is any electrical connection between the case and the enclosure (mounted to 3d printed plastic DIN rail clips)

Could there be some sort of issue with the circuitry of the XYZ probe? There aren't any schematics of it anywhere so I'm actually not sure how the three wires work in conjunction with the magnet (Although I think it may be that GND is connected to the magnet, and when the magnet/router bit touches the aluminum plate it completes the circuit back to the controller?)

Any ideas would be... useful. I'm kind of at a loss here. I do have continuity between the negative terminals on the PSU and ground... about 170 ohms. So I know there is an interaction somewhere.

 

update 1:
I did find what I thought was the issue. I disconnected all grounds from the bus (with it off, but still plugged in to USB to my computer) as well as every connection to the BB because I thought there might be something internal messing with it. As soon as I disconnected the Z limit switch (which I had a modified isolator on as I needed to bring it closer to the frame a bit) the light on the probe went off. Success!

Or so I thought. I started hooking up everything else, basically in reverse order. When I went to reconnect the main line to earth (which does go through a terminal block) the light came back on (with the probe magnet attached to the enclosure mounting bracket.)

So I'm stuck tracking down another GND to earth... I'm a little worried that somehow there is a GND to earth connection in my computer, that's pretty much the only other place I can even imagine there being a short.

Ideas? Other than use wifi lol.