Another plasma component...

Phill

Thank you. I am sure I can make this work for line tests.

I cloned plasmac to the machine control system and it appears that the sim starts up with no errors!

I will look over the new code as soon as I have time. I will start to port it over to my hardware on the weekend maybe Friday. Next two days have a trip to the county seat for old age business, health care is just crazy here by industrialized world standards, and then Thursday to the mainland to drop off parts for powder coating, in between I need to weld on hangers to about 30 pieces.

John

Phill

I updated from your branch. I had a good look at your plasmac.comp this morning and found what I think is a problem, remember I am not much of a coder so I very well can be misunderstanding your code. When ohmic sensing is enabled a check for a shorted torch tip must be done before each probing cycle. It also helps if an error message is posted so the operator knows what the problem is. The torch tip will get shorted when cutting from debris or water. This is especially true if you are using HT Finecut consumables because the gap between the shield and the nozzle is so big. The reason this is a problem with all blowback torches is that internal to the plasma power supply the torch nozzle is connected to the work piece potential between cuts so that when it comes time fire the torch there will be a potential difference between the nozzle and the electrode to make spark for the pilot arc when the connection is broken. On HT torches the electrode is moved and on TD torches the "start cartridge" disconnects the electrode from the nozzle, others I don't know. The nozzle is then disconnected from the work piece potential after a short time or when the main arc is sensed.

I really haven't gotten my head around using offsets to drive the z axis for probing yet. The Z axis raising waiting for the connection to open needs to be much slower than the initial contact probing speed.

John

@John,

I like your words :
I am starting to get a feel for this now so I am going to clone it to my machine controller. I will start porting to my hardware (7i76E & THCAD) shortly.

As soon as I can get an accurate probing figured out I will do some line tests as I can write that code without using any CAM and a new PP.

Today i did a one of a last update to my THC2 controller for parport application's. It's now 99% finished how i want it to be. The update's i do are short update's because there are linuxcnc machine's already cutting every day. For parport i make update's that are compatible.

For the Ethercat THC code, i started from zero 3 weeks ago. This one is a very clever component. It is time ahead in plasma world.

If we keep in mind. M3 & M5 for component logic. It's all in the game. Probing is a component piece so does not have to be mentioned in the G-code.

Keep up the good work !!

islander261 wrote:
The Z axis raising waiting for the connection to open needs to be much slower than the initial contact probing speed

This speed is defined by the offset scale which is (thread period x units per mm)
For a metric machine with a 1mS servo thread this equates to .0001
It only steps one unit per period so this is 60 mm per minute or 2.36 inches per minute

islander261 wrote:
I had a good look at your plasmac.comp this morning and found what I think is a problem

Thanks for that, I totally missed it...
I have just pushed a fix which also includes better error handling for the float switch and the breakaway switch

Cheers, Phill.

Phill

Thank you.

Now I was trying to add to the materials list. Is it true that each material must start with "Mild Steel"? I tried adding materials to the configuration file and they don't show up in the GUI.

During my routine cutting I normally use 2 to 4 different "tools" per piece to very cutting parameters. Right now I do this by directly setting heights and feeds and using M67&M68 to send over other control codes. I have a hal component that decodes M67 & M68 values into machine actions on its output pins. Currently I have hal input pins on my Glade handler files that read the commands values and set the state of the various widgets accordingly. My current Glade handlers use a different method of changing and storing persistent values to a configuration file than your branch uses so I am a little unsure of how to hack it to allow external changes to take effect. I know how to add hal pins to the handler files. When you get some time in the future can you please post a code snippet showing the value of a widget being changed by the change in value on an input hal pin? I really need to control the cutting height and THC enable from my Gcode files (I already have a hal component decoder). My current GUI has a check box that blocks the updating of on screen widgets from Gcode if desired. All of this combined with Pedro's RS-485 control of newer HT plasma boxes will make this into an industrial level branch. Thanks again.

John

islander261 wrote:
Now I was trying to add to the materials list. Is it true that each material must start with "Mild Steel"? I tried adding materials to the configuration file and they don't show up in the GUI.

You can add anything you want as long as it follows the same format. After changing the file you need to go to the config tab and click reload.

I will look into your other request a bit later.

Cheers, Phill.