Another plasma component...

Phill,

I'm looking for where the cornerlock function actually exists. Specifically, where I can find the code that shows what it directs the machine to do if it is enabled. I'm curious as to what exactly it does.

Also, I'm sure I have just overlooked it somewhere but I cannot find where plasmac_run.kerf-width-f is defined. I would have thought it would be in plasmac_run.hal but I don't see it there.

Thanks,
Steven

Search for plasmac.comp and have a look there.
Had not time to play with it for the last two months so i do not know if it there for sure, a lot has changed from then.

Like tommy said it will be plasma.comp if you have the source code. Otherwise, you can look at it in the git repo.

github.com/LinuxCNC/linuxcnc/blob/master...ponents/plasmac.comp

Check out line 965 or so.

I'm looking for where the cornerlock function actually exists. Specifically, where I can find the code that shows what it directs the machine to do if it is enabled. I'm curious as to what exactly it does.

As mkardasi said

Also, I'm sure I have just overlooked it somewhere but I cannot find where plasmac_run.kerf-width-f is defined. I would have thought it would be in plasmac_run.hal but I don't see it there.

plasmac_run.kerf-width is a HAL pin created by plasmac_run.glade for the Kerf Width SpinButton. The -f denotes the FLOAT pin and there is also a S32 pin -s which is not used.

Corner lock is really quite a trivial feature to code. There is a stand alone example I did in the plasma forum here quite a while ago.

Basically you just monitor motion.current-vel and if it falls be low a threshold (say 90% of the original cut velocity), you set a pin to disable the THC until its back where it needs to be. This is one of the things an external THC can't do that sets PLASMAC apart.

Phill,

Because our discussion has turned back to plasmac and away from post-processors, I thought I would move us back to this thread. I've added code to adjust the kerf. Without some real kerf width vs. speed data, it's hard to know if this is really useful or complete overkill. Anyway, here it is. I stayed with your pattern of adjusting all arcs below a certain radius, not just circles.

This will only work with compensated code (no G41.1 or G42.1). I think I could get it to work with uncompensated code but it will require a little more thinking.

There are a couple of other things that Jim Colt writes about improving hole quality - start the lead in at the middle of the hole (maybe easier to set this up in the post-processor than in plasmac) and have the torch turn off exactly where the hole started while still moving (basically continue the arc a little further). I'll see if I can figure out how to do one or both of these. My thinking is that putting these things in plasmac makes a lot of sense as they are clearly plasma specific.

Steven