Gmoccapy - A new screen for linuxcnc

Hi Todd
Norbert is correct about the Mesa card. You can use many variations of THC control.
With the Mesa THCAD I use two components. The first one look at the voltage that comes from the THCAD and compares it to the THC Voltage setting on the screen. It will output a UP/DOWN signal that is coupled to the THCUD component ( already in the distro's). This component will control the Z axis to follow the material at the correct height. This method requires you to make use of Gcode to control the torch actions.

There are many settings on the Plasma screen but not all are needed for every system. We have included many parameters in order to allow users to create a very comprehensive control system. Many systems use a hardware THC and then it only makes use of Gcode and a simple up/down signal setup. You could also do the whole THC control in a component if you use just the THCAD. It al depends on how much effort you want to put in.

I have been busy with a very comprehensive THC controller in software for some time (to many concurrent projects ) and I hope to finish it soon in the new year as my own plasma machine will be done by then. I will make use of all the settings on the plasma screen as I am only using a voltage input similar to the THCAD. The rest is done in the component without using Gcode. The M3 is remapped in order to allow me to control the sequence of events of the plasma.

Let me know if you need help on setting up the system. I suggest that you read up on the wiki about plasma procedures. I have put a lot of information on there that could be useful in understanding the process.

Plasma wiki

Hello all. I seem to have gotten an error after some updates that I can't track down, here are the 2 error sceens I get when trying to open. I have made no changes to the python file in these areas.

Hi Johnny
What did you edit or change and what release are you on?

mariusl wrote:

Hi Johnny
What did you edit or change and what release are you on?

Hello Marius,
That's the odd thing. 2 weeks ago I had everything working (On 2.6.4-Sim) and was about to move on to HAL file edits to my actual hal files at my machine. Fired up my virual machine yesterday morning after not having worked on anything for 2 weeks, did normal udates, ended up on 2.6.5-Sim) then got those errors when attempting to open my GUI. No files were changed in that time. I reverted back to 2.6.4 and still get the errors. I was just trying to see if it was an isoloated problem, and it must be. Thanks for the reply. I'll just start troublehsooting.

mariusl wrote:

Hi Todd

Let me know if you need help on setting up the system. I suggest that you read up on the wiki about plasma procedures. I have put a lot of information on there that could be useful in understanding the process.

Plasma wiki

Thanks for the offer to help, I think I'm going to need a little!

Is there a config folder from a working Gmoccapy plasma machine that is using a THCAD for THC control that I could look at? I think I am just getting hung up on the differences between a demo setup and an actual working configuration. I don't have a problem using G-code for for probing, delays, and heights, I do most of this through my post processor now.

Thanks again,
Todd

What I would suggest is that you use Pncconf to do a standard 3 axis setup and then just add the THCAD input. It is a frequency that you read with one of the encoder inputs on a 7i76 or the like.
You convert the count to a voltage and put that through a component like this:

This is then connected to the THCUD component. look at the Gcode examples that I gave in the wiki.

Pncconfig does not have gmoccapy or gmoccapy_plasma as one of the options. I have the configurations working in axis including the THCAD to voltage. The axis configurations do not work when I copy them to gmoccapy_plasma. That is why I was hoping to see a functioning config folder. I understand it will not be correct for my system, but I may be able tell where my mistakes are by comparing to a config that is known to work.
For example, I am assuming that I should be copying the contents of my axis.hal to the core sim.hal in gmoccapy, but I am only guessing.

Is there a copy of gmoccapy plasma that is not setup as a Sim?
Is there a pncconfig that will configure gmoccapy plasma?


Thanks,
Todd

OK,
I'm making slow progress, I have motion, I can read torch voltage, I have a post processor that is working all of the probing, timing, offsets and firing of the torch, but I am stuck on how to use the component that converts the voltage to a up down signal.

Todd

Hi Todd
Did you compile and install the component that I attached before?

You have to connect the set-voltage-in to the plasma screen tip voltage and then connect the arc-voltage-in to the analog voltage that you got from the mesa card. Now use the up and down signals to connect to the corresponding signals of the THCUD component. Do a "man thcud" to see what is required.

In short, the comp I attached will look at the incoming arc voltage and compare it to the set-voltage. If it is too hight it will set the "down" signal and if it is too low it will set the "up" signal. There is a signal called voltage-window. This allows you to set the window size for the comparator. The THCUD component will use these two signals to adjust the Z axis up or down and thus follow the material.

This thc200 component just replaces a piece of hardware that gives up and down signals. THCUD is usually used in conjunction with a hardware voltage comparator. thc200 is used when you just have an analog signal and simulates the hardware. For the rest you use THCUD as the help describes.

Hope this helps a bit. I should really do an example config for the wiki soon.

Let us know if you dont come right and I will try and do you a sample config.

Its a bit sad to use an up/down THC comp when you have the actual arc voltage.
Better dynamic performance would be possible by using a real PID loop
for the torch height.