(Solved) PlasmaC new config Z going up when probing

phillc54 wrote: That seems to show the down led on just before the crash.

It would be interesting to see a plot with the arc voltage and the up and down leds

I agree... if I knew how to use halscope. I've played with it to extreme levels of frustration, and while I've gotten it to do what I think I wanted it to do before, basically it was an accident... I also couldnt find a clear and concise video/set of instructions. It's on the "to conquer" list I guess. Seems like a useful tool.

PCW wrote: Looking at the cut, I think it runs into the material before the THC is enabled

Maybe, but how would I avoid that, other than giving it a longer land to get going on? I had the delay all the way down to like .3s. I guess those are my two options? Maybe tomorrow I can make some more test cuts with a longer flat land before the steep changes in height. See what happens.

Shortening the THC enable delay (say .100 second) and having a lower
corner lock threshold (50%?) would shorten the distance from a static start
to where the THC gets enabled, but as I mentioned before, starting motion
from a stop on material with a Z ramp and expecting the THC to work instantly
is expecting the THC to work where its been explicitly disabled by other features...

I don't know if you noticed but I cheated and started downhill so the torch did not run into anything
You need to have a flat area to cover the movement in that first 0.3 second .

rodw wrote: I don't know if you noticed but I cheated and started downhill so the torch did not run into anything
You need to have a flat area to cover the movement in that first 0.3 second .

And exceed the corner lock velocity, whichever happens last...

PCW wrote: Shortening the THC enable delay (say .100 second) and having a lower
corner lock threshold (50%?) would shorten the distance from a static start
to where the THC gets enabled, but as I mentioned before, starting motion
from a stop on material with a Z ramp and expecting the THC to work instantly
is expecting the THC to work where its been explicitly disabled by other features...

Are you autosampling torch volts? I think that might be best. That way if the torch voltage has not settled back to cut voltage by the time the THC kicks in, you will end up with a torch voltage that is higher than it should be so the torch will be higher. There really is no substitute for learning how to use Halscope and observing some cuts with THC disabled so you understand the importance of the THC delay and how long that delay should be.

So I made some videos of my problem. I got a piece of plate with a nice flat start, and then added a curve. The problem is, it never even makes it to the curve. I am 90% sure the THC is adjusting in the wrong direction. I kept getting "float switch tripped" errors, which is odd to me because the material doesn't really move. So I figured I'd brute force things and clamp so that my float switch couldn't trip. That causes the torch to fall off of the magnetic break away, you might think it runs into the material, but upon closer inspection what's going on is the torch bottoms out, and it continues to adjust in a downward direction, causing the torch to pop off of the mount.

I'm not entirely sure what's going on here, so I appreciate any feedback. From what I can see in the videos, it just keeps adjusting down.

PID was at 20
Delay was at 1s for the first video, .5s for the second video.

THC1 Machine
THC1 Screen


THC2 Machine
THC2 Screen

I appreciate any insight.

Try it with Use Auto Volts enabled.

Edit. I expect your actual cut volts is higher than what your material setting is.

phillc54 wrote: Try it with Use Auto Volts enabled.

Edit. I expect your actual cut volts is higher than what your material setting is.

Well, a brilliant man you are. Auto volts got me there. In conjunction with raising the torch another .070", so that the cup doesnt touch the metal. Which wouldnt be possible without the auto volts. Good stuff! I think I'll cave and just keep using auto volts.

THC3 Screen
THC3 Machine

One thing... Linuxcnc froze during the one cut, and I got a weird error. I accidentally closed Linuxcnc.. Any ideas where that log file is stored?

I think the linuxcnc log is restarted each time you run it, they are in your home directory.

If you want to use the Voltage from the material file you need to ensure:
1. The voltage on the material file is correct for that cut height.
2. The voltage that PlasmaC displays is correct.

You do need to do a lot of test cutting to ensure your settings are correct. Most folk go with Auto Volts.