Doing Ohmic probing and THC with one THCAD10

Hi all,

I'm new to the forum and am currently working setting up my plasma cutter. I'm using an MPCNC currently to learn a lot and get a working machine first and then will invest money and time in building something proper.

I have a 7i76e and a THCAD10 and the arc voltage seems to be read properly at this point. I also have a dual optically isolated relay plate. Both relays have one normally connected and one normally open contact. I was thinking the following:

If I connect the THC voltage divider from the plasma cutter (Hypertherm 45XP) to the normally closed pins and the common contact to the THCAD10, I would read the plasma voltage during normal operations.
To the normally open contact, I would contact an isolated 24V power supply and basically mimic the diagram as laid out in linuxcnc.org/docs/2.8/html/plasma/plasma..._with_a_mesa_thcad_5 (Hypersensing with a THCAD5). I then would need to change the resistor to match the 10V, that's clear.
Now I need to "just" get the linuxcnc config right so that it triggers the relay during omic probe and reads from the encoder that I also use for the THC.

To me this (if it works) would have the following advantages:
- only one THCAD needed, thus simpler build and money saved
- no special bit file (that is hard to find - I didn't find it yet for the 7i76e) needed

I suppose I'm missing some point, if not I'm happy to give it a shot and try it out.
Thanks a lot for your input!

Klaus

You should be able to do this but the OPTO relays must have MOSFET type
outputs to switch the low level analog signals without voltage drops.
Also beware that the probe input can have fault voltages up to full
plasma voltage so the relay output must hold off these voltages.

BTW the 7I76E bitfile for plasma is 7i76e_7i76x1_Dpl.bit

Hmm, I have mechanical relays with optocouplers, this one: smile.amazon.de/gp/product/B01K9S6Y16/re...07_s00?ie=UTF8&psc=1
(I'm from Germany, so sorry, only have this link). Looking at the data sheet for the sla-24vdc-sl-c it should be isolating enough for any voltage that the plasma cutter could create. WDYT?

Thanks a lot for the bitfile mention - maybe we should add this to the official documentation.

Relays should work but if they do not have gold plated contacts you may have issues with the low currents switched.

Ok, thanks a lot for the input so far!
I'm trying to set this up and I'm baffled... here's what's happening.

I've basically replicated the setup that is laid out here linuxcnc.org/docs/2.8/html/plasma/plasma..._with_a_mesa_thcad_5
I'm not using the AC relay, I'm using a MeanWell HDR-15-5 power supply on the THCAD10. What happens is that if I connect any side of the powersupply (+ or -) to the positive side of the THCAD, I begin to get weird readings. That's right - without the negative pin being connected at all. The arc voltage starts going from -4 Volts up to 200 and back. In the HAL-meter this looks a little more chaotic, but I see the values jumping.
I've tried with a HDR-15-24 then and the situation is the same.
If I connect both sides, + and - then I get the correct reading of 102 Volts (Voltage divider value is 20:1, so 5V input = 100V roughly). If I take away the negative pin, I see the jumping readings again.
How's that possible with a double insulated power supply?
I also tried to ground all possible things but nothing helped.

Any input is appreciated!

I'm beginning to doubt that the THCAD with it's super-sensitive input is a better choice than using the relay method... and I'd love to use the THCAD for all the good reasons.

A double insulated power supply may have considerable AC hum potential
until you have a reasonable load for sensing and proper grounding (at least
to the table), you would expect to pick up a fair amount of AC hum.
The other thing that can help is connecting the THCAD input shield to the + input