QTPlasmaC mode2 - valid arc lost

I have eliminated the line with "torch-off-probe".

Oddly, QTPC is not reading my config values (in custom.hal) for "db_arc-ok.delay" and "plasmac.arc-lost-delay" :

I have noticed something strange.

When IHS occurs and torch is fired,
my external THC is set to wait for a delay of 1.5s before sending UP/DOWN to B.O.B.

As seen in the video below, in the right/down corner, THC DOWN is ON the moment the ARCOK is ON. And Z is diving until it hits the material. The problem is the external THC is NOT sending DOWN at that moment (I checked visually, don't have 2 cameras to record both QTPC and external THC).


I guess that when it hits the material, the plasma source extinguishes the ARC (my guess) and THC Arc OK is GONE from the external THC.

Any clues?

I have just confirmed with a multi-meter that external THC is not sending DOWN signal while THC DOWN Light (on QTPC) is ON.

My external THC is fed by a notebook source at 19V (specs are from 12 to 24v).
The DOWN input from B.O.B. (or DOWN output from external THC) reads 10V when not activated and 200mV when active.

The same happens with external THC UP / ArcOK outputs.

By "Disable THC in LinuxCNC" you mean the button at QTPlasmaC interface?

Should I keep all the wiring between external THC and BOB ?

I've found an old topic you give lots of tips, will extract everything I can from that
 

Thanks tommylight, I have followed your advices and things are becoming clearer now.

My THC Arc-ok closes the Chinese BOB input when the sensor is active (it goes down to 0 Volts), I thought it would translate into a TRUE when the sensor is active but it is quite the opposite.

So I have inverted all the UP, DOWN, Arc-ok inputs and it's looking good at the QTPC GUI. Next I'll try with the Plasma Source On.

 

I think QTPC THC docs could have an more through approach like it was done with "Homing" (it was a breeze to learn that).
I'm not complaining, you guys are awesome, thanks.

I would go section by section and make notes about improvements, not sure where you host that (github?), but that would be a work apart.

The general idea I'm talking about is that PlasmaC documentation is very complete, few things missing if any, the problem is that some complex topics (like THC) deserve a more didactic approach. Things like:
- illustrations and images instead of text only;
- real life examples;
- explanation (or pointers) to related concepts and pre-reqs;
- what can go wrong on most common mistakes;

Anyway, I see the docs improved a lot overtime, and I see it is slowly getting more like a book than a dictionary. I hope I can contribute with that as soon as I leave the noob arena.