THCAD 10 to Herocut 55i (1:1)

Is my idea to tap into the raw voltage a bad idea, or shouldn't it work?
My thoughts it that then i would "calibrate" the voltage to be spot on?

You don't need to calibrate the plasma, you just need to set the THCAD parameters up correctly so to rush in and change things now might be a step backwards.

lets see solve for X (volts) from thcad manual, 
(x-10)/100 = 2.24 M
x-10 = 2.24m*100
x-10 = 224
x = 224+10
x =234 volts

So 10 volts in thcad = 234 volts
so your divider is  23.4:1
(234/10)

So what are the 0 volt and 10 volt numbers on your calibration sticker ?
Have you got the link on the thcad set to 1/32?
 

Do not open plasma.
The voltage does not have to be spot on, it will work even if voltage on screen is way off.
Set the scale and offset to what the calculator says, start using the machine.
If the voltage is way off, adjust the scale a bit.

Thanks for the help.

0v 118,2kHz
10V 938,9 kHz
Yes i have it set to 1/32.
I thought that i hade to put the scale to 2240000 :side:

Ok, we will use Herz here:
Offset =    118200/32  = 3693.75
0 volt to 10 volt frequency range = 938900/32 - 3693.75
= 29340.625 - 3693.75
= 25646.875 per 10 volts
= 3564.6875 Hz per volt
But Plasmac uses volts per Hz so take the reciprocal 
0.000389911051541

So your data =
offset = 3693.75
scale = 0.000389911051541

Lets do a test. Say the frequency on the encoder was 23693.75
deduct offset = 20000
THCAD internal voltage 20000 * scale = 7.798 volts
Torch voltage = max scale (234) * 7.798/10
= 182 volts
I would expect to see cut voltages of between 90-130 volts.
Its always best to take the time to understand how this works.
If you use halshow to find encoder.n.velocity when not cutting it should be hovering round the offset of  3693

To absolve some possible confusion, I feel compelled to point out there's a typo below (change red):

rodw wrote:

= 25646.875 per 10 volts
= 2564.6875 Hz per volt
But Plasmac uses volts per Hz so take the reciprocal 
0.000389911051541

So your data =
offset = 3693.75
scale = 0.000389911051541

That's true, for a voltage divider of 1.

For a voltage divider of 23.4 (value calculated in post above), the values are as follows:


offset = 3693.75
scale = 0.00912392


Another source for the calculations: linuxcnc.org/docs/devel/html/plasma/qtpl...c.html#qt_mesa-thcad
And a handy calculator for the values: jscalc.io/calc/NTr5QDX6WgMThBVb

Thanks, 
Now I know why I still scale my torch voltage with a component I wrote years before plasmac was around that does all the maths for me. I set my scale = 1 and offset = 0.

I can't quite remember the issue but very early plasmac did not let me enter the required values due to an error in the field validation limits.

0.000389911051541 * 23.4 = 0.00912392
 

It’s working!