Question about THC parameters

Hi all,

Our plasma cutter is a bit temperamental and we are chasing the cause of that. For some reason that we can't work out, the machine is cutting along nicely, but then all of a sudden it lifts torch up a couple of millimeters, and of course that throws off the controller and the torch lifts itself further up until all it. Same thing happens sometimes in reverse, the torch all of a sudden dives in the material for apparent reason. Mind you, these symptoms occur at random times. When these events occur we often resort to using IGNORE ARC OK, but of course the THC doesn't work then.

We suspect it has something to do with the parameters of the THC, but I can't find anything in the user guide that explains what some parameters do. Can somebody shed some light on what these parameters do and how they affect THC functioning:

• Treshold                (currently set at 1 Volt)
• Speed (PID-P)        (currently set at 10)
• VAD Threshold      (currently set at 80%)
• Void Override        (currently set at 80%)
• PID-I                        (currently set at 20)
• PID-D                      (currently set at 20)

Any insight will be greatly appreciated.

Cheers, Peter

Threshold: the amount the arc voltage is allowed to vary above or below the target voltage before THC is activated.

PID-P, PID-I, PID-D : are basically standard PID parameters, most folk use the defaults of 10, 0, 0. You may need to increase Speed (P) if your Z axis is a bit slow. linuxcnc.org/docs/html/motion/pid-theory.html

VAD Threshold: when the current velocity falls to this percentage of the cut feed rate then THC is disabled, it re-enables when speed has recovered to 99.9% of the cut feed rate.

Void Override: An adjustment for kerf crossing threshold voltage. For some folk(including me) this works, for others it doesn't. There have been complaints about it being wrong but I have asked several times for data to help in changing this but none has been forthcoming (it hasn't stopped the complaints though) . I am thinking of removing this totally from QtPlasmac and folk can sort out their own system of kerf crossing, I am past caring...
 

Hi Phil,

Thanks for that information. I will have a play with the PID-P setting, that might reveal some feedback. Our Z-axis seems to be responsive enough when the machine is cutting normally, but we might be just on a threshold where it's a fine line between normal cutting and throwing it out of whack.

Cheers, Peter

From symptoms i would venture a guess that you are using a quite old version of PlasmaC, it had a knack for disabling THC at random, especially on longer cuts, so the issue would be dependent on if the material is lowering or raising.
And update will fix all that, if i guessed right.
 

Hi Tommy,

Thank you. We were using QtPlasmaC on a 2.9 build, but I am in the process of upgrading to Buster (which is causing some headaches). See what happens if we get it to work.

I have a question I'd like to ask. Does the number we specify in PID-P mean the percentage of Z axis velocity(i.e 10%)? My Z axis is at 4500mm velocity now so does 10 in PID-P mean my THC speed will be 450mm?