THCAD Environmental Fluctuations

Agreed Tommy, I even bought an oscilloscope last time and concluded the noise was entering via the mains ground!
A mains EMI/RFI filter stopped my problem immediately.

Huh, maybe I should try watering my grounding rod when TIG welding so I don't have to go reset all my house GFCIs afterward. Grounding to my plastic water pipes not likely to help.

Hmm, I shared this with a US based Colleague overnight and he said:

I am definitely not a professional electrician but my first gut reaction is that if you see a ripple from a ground, you may have a ground in neutral tied together in a sub panel somewhere

As it turns out here in Australia where both thefabricator03 and I live, that is required under our electrical code where we have adopted the MEN system (Multiple Earthed Neutral).

Ref: electricians-success-academy.com/learn/what-is-men-system

So this begs the question. What should we do as best practice to wire THCAD's?
Should we even attach the THCAD to frame ground?

Actually for most systems, the THCAD shield should probably
be connected to the low side of the divider (the +in)

PCW wrote:

Actually for most systems, the THCAD shield should probably
be connected to the low side of the divider (the +in)

Hmm, thats a bit contradictory when the manual says

The THCAD has an input shield that should be connected to frame ground on most systems. The input shield is the ground signal to a common mode RFI filter on the THCAD

Can you elaborate a bit please Peter?

For the lowest noise, the shield should be connected to the low side of the divider
(which should be closest to the "work" connection) But this may not be the best choice
from a safety perspective

PCW wrote:

For the lowest noise, the shield should be connected to the low side of the divider
(which should be closest to the "work" connection) But this may not be the best choice
from a safety perspective

Thank, that definitely makes sense on the Ohmic sensing circuit as the isolated power supply is double insulated and hence has no connection to earth. But it does make me wonder about the arc voltage that we know affects the voltage seen by the THCAD in the circuit.

I'll try and integrate a RFI filter into the mains input on my test bed and report back. I think it will solve the problem.

Well I cut a bracket so I could use an IEC connector including RFI filter, power supply and fuse.


My associate said this this morning.

I kind of worry that adding a filter to mains will fix it. If the neutral/ground connection is what's causing the problem, then the filter won't touch it.

And he was 100% right. It made not one jot of difference!

So that leaves us with the two ideas from Peter and Tommy.

I don't think the shield connection would make any difference for ohmic sensing as there are no shielded cables in use on my table. Plus the power supply is double insulated and has no ground point. One wire is bolted to the table near the enclosure at the star ground point and the other runs up unshielded to the torch tip.

So doe anybody have any feedback on Tommy's method?