Cool plasma machine

QtPlasmaC supports 5 axes so if you feed it the right gcode then it should be fine.

EDIT:
Thinking about this a bit more...
We hide all Z motion currently so you would need to rely on THC for all Z motion.
It would be possible to add to the filter code so that a magic comment in the gcode would prevent Z motion from being commented out.

I have been working on such a thing for quite some time now ... well more like been thinking and testing some ideas on how to make the big rotating part, very hard to find anything i need around here ...
So far i can dispense some info that i figured out the hard way and some general info:
-round tube machines are easy up to roughly 200mm, above that and the machines start getting very complicated and very heave really fast, had a request lately to build one at 12 meters by 0.5 meters so i asked what would the weight of the pipe be, it came out at over 1 ton ...
-square tube has to be 4 OR 5 axis machine, depending on if you also want beveling, but it can be 3 axis for up to 60x60mm with a fast THC, maybe can be pushed to 80x80 but i did not test.
-square - requires very powerful motors to rotate and push/pull the material so any mishap ends up badly for the machine - this is the part that is never ever shown in videos
-square - the only tricky part is the front holder/clamp/centering/rotating/free-moving/.....add more here !
-
So if by any chance you end up jumping into it, i would really appreciate the heads up.
And good luck, this really requires some luck.
P.S.
I already went through two hardware iterations of the clamppy/rottary/whatever/thingie that failed miserably!

phillc54 wrote:

QtPlasmaC supports 5 axes so if you feed it the right gcode then it should be fine.

EDIT:
Thinking about this a bit more...
We hide all Z motion currently so you would need to rely on THC for all Z motion.
It would be possible to add to the filter code so that a magic comment in the gcode would prevent Z motion from being commented out.

We do have a magic comment for that currently, I had asked for it a while back (before QtPlasmaC) as it makes using a spring loaded marker on the table far easier:



But I think that's only half of the solution and we would also need a way to either:

1. only have the automated Z movement be the probe to set the top surface of the material, and then follow the Z movements in the G-Code from there.
or
2. let the user set the top surface manually in these cases, and do no Z movements whatsoever on an M3 command.
or
3. #2, but instead of M3, map a different M code to turn the torch on and off when following Z movements

Anyhow, I'm literally spit-balling from the top of my head, I'm sure it gets way more complicated than that.


Fun video though, I wish they showed how they made the waterfall of coolant through the tubes/pipes being cut. Interesting approach.