THC Delay in material file

Hi all,

We are wondering if it is possible to set or override the THC delay in a material  file ?

We cut a wide range of material thicknesses, and we have found that on thick (25mm) material we have to increase the THC delay in order for the arc to sufficiently stabilize and get a good voltage reading. If we could set the THC delay in the material file then that is something we don't have to change manually and avoids operator errors.

Cheers, Peter

tommylight wrote:

linuxcnc.org/docs/devel/html/plasma/qtpl...qt_material-handling

Tommy, thats not helpful. He wants to change THC delay not Pierce delay for different materials. AS it stands, it can't be done.

Which leads me to a theory I have after observations after significantly upgrading my machine's acceleration.
Perhaps the THC delay is the wrong way to go about it.
Maybe we should be enabling the THC when it gets to cut speed or constant velocity.
The hypothesis is when the current velocity is below the cut speed and accelerating, the arc is unstable and it stabilises when it gets to speed.
But it seems Peter's experience disproves the hypothesis.
If the hypothesis was correct, a component could be written that varied the THC on delay based on cut velocity.
I don't think the guys developing the software have the ability to cut thick materials with their plasmas so they don't see some of these issues.

I can certainly see that adding a THC delay parameter to the materials file that overrode the default if present would be a useful feature.
I think also adding parameters for material and material thickness to the materials file would also be useful for future enhancements. 

 

Hi all,

Tommy's idea would work for us, we can add something to our post processor to include a line like "TD=2" or similar. What is the proper format for that ? I can't seem to find anything in the user guide to that extent.

Cheers

I don't think there is a gcode option for thc delay. If there is it will documented.

snowgoer540 wrote:

EDIT: It was a bit ambiguous but the change was pushed 13 days ago with this commit: github.com/LinuxCNC/linuxcnc/commit/9486...171f45821da5aa4d4096[/i]

I think this is a good change as it also allows for different acceleration rates between machines. I'm glad you mentioned it.

Monitoring with halcope will help understand behaviour and set a suitable delay.

Looks good!
What thickness material?
Whats with the voltage spikes? Are you testing kerf crossing?