Upgrade problems

Hi Phil,

I have set the Arc OK to 400 volts (was at 200) and that got it going ! Then I realized that the actual arc voltage shown was about 140 with the torch off, for some reason that has changed. When I change the voltage offset to read the arc voltage 0 at idle, then the torch doesn't fire, not even a pulse. This is another thing that hasn't happened before ?

Is there some documentation out there that describes how PlasmaC works internally, that might make it easier to trouble shoot ? At the moment I see a lot of variables in the config panel, but I can't seem to find any description of what figure does what, or how it impacts the controller, That makes it hard to understand how the thing works. Any ideas ?

Cheers,

The two that are relevent here are Voltage Offset and Voltage Scale in the Config Panel.
linuxcnc.org/docs/2.8/html/plasma/plasma...de.html#config-panel

They should not have changed with the update and should be what you calculated here:
linuxcnc.org/docs/2.8/html/plasma/plasma...uide.html#mesa-thcad

How is the thcad connected to your machine?

Hi Phil,

It turned out that I had to remove the '-' sign from Voltage Offset and Voltage Scale, as we had it previously working. Not sure why the sign had to change ? We are using mode 0, and I have to set Arc OK Low at 0 to get the machine to cut. It means that the Arc OK is always on, but in mode 0 it's probably not an issue.

The THCCAD signal is fed into a breakout board to a parallel port, with an encoder thread in hal. That part has worked right from the start, and without any problem so far. Never had to touch any setting on it.

I am going to try and see if I can feed the Arc OK signal from the Hypertherm in the controller. Previous attempts at that failed, and I think the signal from the Hypertherm needs debounce to be accurately read by the software. We had the same issue with the float and home switches, lots of problems at first and with debounce they work no worries at all.

Cheers, Peter

Yes, it is best if you have a Transfer/Arc OK in the plasma machine to use it and switch PlasmaC to Mode 1.

You should (in Mode 0) have some voltage level for ArcOK Low as you may get false starts otherwise.

PlasmaC does normally debounce the float, ohmic and breakaway switches automatically, you may just need to increase the values.

Hi Phil,

Thank you for your help today, much appreciated !

I have tried to use Mode 1 with an Arc OK signal from the Hypertherm, but I get an error "breakaway switch activated" when the torch is touching off. A bit strange, we don't have a breakaway switch hooked up and I have commented out the line to do with the breakaway switch in the plasma connections hal file. No clue to why that is.

I have set everything back to mode 0 and set ArcOK Low to 0 and it seems to work, anything else gives problems, so another thing that baffles us. I suppose we stick to what we know works, and let everything else be. Too much time consuming trouble otherwise.

Thanks again,

Cheers, Peter

CFE wrote:

I have tried to use Mode 1 with an Arc OK signal from the Hypertherm, but I get an error "breakaway switch activated" when the torch is touching off. A bit strange, we don't have a breakaway switch hooked up and I have commented out the line to do with the breakaway switch in the plasma connections hal file. No clue to why that is.

Are you connecting the Arc OK signal to debounce.0.1.in by any chance?

Just thinking about some of the strange issues that you are seeing makes me think that maybe you have a grounding problem.

Hi Phil and Tommy,

How to find out if there is a grounding problem ? We are not electricity experts, but we are using proper shielded cables from the controller to the motors and switches, and run the earth clamp from the plasma cutter as close to the material we are cutting as possible. And the plasma unit is a Hypertherm 85, we assume that is a good quality machine and we like to think that for instance the Arc OK signal from that machine shouldn't cause issues. Are there any guides as to good grounding practices ?

Cheers, Peter