7i96s Phantom Voltages

OK so I'm finally getting to the let's get some motion point of this build and I'm pulling a phantom voltage off the Proximity switches. It's generally under 1 vdc but enough to just barely illuminate the little LED and throw a random flickering signal in halshow.
Proximity setup is Powered from a small 12vdc power supply stepped down from 48vdc. Runs in a 12core 22awg shielded cable to a box mounted on the back of the gantry. from there the wires are distributed via aircraft plugs. 
Initially I had all grounds 48, 12, 5 vdc running to the same block. changed that so each power source has its own ground even though it all eventually ties back to the grounds from the 48vdc power supply.

Anyone know what might be causing this. I have some ideas but hoping I'm wrong.

1V DC will not trigger the inputs (they need about 3.5V minimum) but if you measure
1V (average) you probably have large but short duration voltage spikes coupled into
the inputs somehow, This may be either coupled capacitively or inductively from things
like motor drive wires or  from ground loops. Is the shield connected to the input common?
Servo and step motors often couple lots of HF noise into their frames so the machine frame
AC grounding is important and frame --> Input common connection needs a good AC GND
 

I might have found the issue. it appears these cheap Prox switches from Amazon are sensing metal around it and allowing a small voltage to pass through. if I lay the switch down it gives a small reading if i hang it from the gantry no metal for a good 7-8 inches around it nothing appears.

Ive used these proximitys for years running under Mach4. So never really thought they would pose an issue.

With out breaking the bank is there a go to for Proximitys for plasma tables?

Looks like the issue is going to be these prox switches side sensing. if metal is to the side of it with in around 20 - 30 mm its sensing at around 1.68 vdc.

I installed a limit switch in place of the prox and no issues the input basically goes to zero until tripped.

Ill Look into the cap I have yet to fire the plasma yet so things are still going to be interesting in what might spike or create issues.

Replaced prox switches with PNP NO Switches and issue appears to have left. I now seem to have an issue in the motors where once they are moved they bounce or rather bump fwd and rev until the oncreen power button is pressed. Im guessing its something to do with my motors not being dialed in yet.

All issues here have been resolved. the cheap amazon LJ12A3-4-Z/BY Thick head proxies PNP NO for some reason send signals when close to metal regardless of the specs written. NPN NO Switches LJ12A3-4-Z/BX eliminated those signals.

For the motor Jerkiness at idle it was apparently in having my servo thread set too high or the watchdog timeout too high. Im guessing the first one as I'm not certain i set both back to their default values

I ran through PNCconfig to reset everything so I'm not sure if that made changes for me and I didnt see them, all I know is the jerkiness of the motors at idle went away after that change. I just did a dry run with one of files using the plasmac post processor in sheetcam and seemed to go well. Well except my float switch it did a touch than lift than a drop and than lift again Ill need to dial that in a little better as during the dry run it was about 1/4" off the material.

I'm still in need of reading up on the THCad2 for height control. I Asked a few questions last month and need to go back and touch base on those. cant remember what all i need to add or do to ensure it is going to work properly. but that is probably a few days out still as i need to weld in a few more supports for my waterpan.