PlasmaC torch starting problem (Solved with help from the famous three)

rodw wrote: And Arc Fail timeout of at least 2 seconds due to laggy torch start.

Rod I do thank you for looking into this. But I am not sure what you are asking me to do, and how to do it.

Is the code aimed at me or Phil ?

Phil has suggested some options here:

So there a maybe a few different ways around this.
install a current sensor in the torch lead and use that for an Arc OK signal
I could add a Arc OK on delay so the arc voltage needs to be within limits for nnn amount of time before Arc OK is true
I could add a Arc OK off delay so the arc voltage needs to be outside limits for nnn amount of time before Arc OK is false
I could add a Arc OK High inhibit so Arc OK won't go false if the arc voltage is too high

I don't want to meddle in the dark and break the system.

I did see Clive's post about 20 minutes after he posted it and as usual i went on to reply to it, but, as i was looking at his graphs something does not add up, well a lot of things.
To be clear, i went to reply the same thing all of you did reply, but, and again a big but, that does not seem to be the case.
The first thing is the noise, to much of it at 50Hz is a sure sign of grounding failure, or a miss wiring.
Then there is the instant torch off after an arcok, normally assumed to be from to much voltage, but it looks more like failure due to a spike in voltage, and not just the torch voltage.
@Clive,
Does the plasma still have hand trigger ? Can you do a voltage trace by activating it by hand while cutting something, once while the THC shows 0V, and once while it is still fluctuating between 0 and 20V ?
I would like to see if there is a difference in voltage between two readings.

rodw wrote:
Thats not 100 ms! Its one second.

The division is the minor tick. 10 minor ticks per vertical line.

That doesn't make sense to me, so in the first of Clive's Halscopes the sawtooth is 5Hz and the arc voltage is around 1400V or am I missing something?
I have always thought the full length line was the division and the minor ticks were a subdivision.

tommylight wrote: Then there is the instant torch off after an arcok, normally assumed to be from to much voltage, but it looks more like failure due to a spike in voltage, and not just the torch voltage.

That is my reading of the plot as well.

Clive S wrote:

rodw wrote: And Arc Fail timeout of at least 2 seconds due to laggy torch start.

Rod I do thank you for looking into this. But I am not sure what you are asking me to do, and how to do it.

Is the code aimed at me or Phil ?

I don't want to meddle in the dark and break the system.

Well it is something you could do. Just keep a backup of the plasmac.comp

If you have a run in place config its just a matter of editing the comp and then going halcompile --install plasmac.comp
For a deb install, I think you have to use sudo.

As I said the Plasmac algorithm is just plain wrong but Phill will have to support existing users.
Either that or set the high volt to 1000. If the GUI does not support that, then unlink the plasmac pin (find it by typing man plasmac.comp) and set it to what you want with setp post gui

The noise is not such an issue as the signalis fairly clean when cutting

@Clive
As a test could you unlink plasmac.arc-ok-high and setp it to some ridiculously high value.
Edit: you could do that in Hal Show with "unlinkp plasmac.arc-ok-high" then "setp plasmac.arc-ok-high 10000"
Hopefully it will get you cutting so we can see some more halscopes.
Could you also include the pin plasmac.state-out.


@Tom
How does a Proma determine Arc OK?
I believe they don't have a window, just a threshold, but I don't really know.
Also do they have issues with spikes similar to this?