Puddle Jump Height and Puddle Jump Delay Values

thefabricator03 wrote: Phill,

Please see attached for the Gcode for the 20mm base plates I was having problems with.

When I pierced the holes the molten metal created a line of metal that went over the outside contour of the part. When I was cutting the outside of the part at cut height the torch would hit the line of molten metal.

This is a bit crude but seemed to work ok in a sim so it should at least prove/disprove the theory. If you want to try it then you will need to:
Rename your existing plasmac_gcode.py link in your config directory for safekeeping.
Copy the attached file to your config directory.
You would need two tools/materials set up for this, the original would be used for piercing and a copy with the pierce delay minimised would be used for the cut.
If you have #<pierce-only> = 1 anywhere in your GCode file the GCode parser will output a piercing only file.
So you could have two original GCode files, one with the pierce command and one without or you could have one original and save a new file after the GCode parser has output it and us that directly.
If it goes tits up just delete plasmac_gcode.py and rename your link back to it original name.
I hope I have explained it well enough...

docwelch wrote: .... Having read a few horror stories about first cuts going very bad due to the slag created by the pierce, I didn't want to repeat those mistakes...

I have direct experience with those "first cut" horror stories. I had just purchased a brand spanking new HT PM85 with which I would be cutting 3/4" steel plates for my client's structural steel fab business.

I received the new machine via UPS and promptly unpacked then hooked up said machine to my table with a test piece of material loaded. Everything was set exactly according to the HT manual. I pressed "play" then watched as my brand new torch and about $70 worth of parts turned into molten slag.

The pierce height prescribed in the manual was obviously too close to the material because the resultant volcano shot out and up the side of the shield and the retaining cup. I stood in shock and by the time I managed to hit Estop, the electrode, the nozzle, the shield and the retaining cup were destroyed.

phillc54 wrote: I did look at "wiggle pierce" and "ramp pierce" some time ago but I seem to recall Jim Colt mentioning something along the lines of HT recommending against it as their torches are designed for stationary piercing and also that having motion while piercing increase the amount of molten material blowing back in unknown directions. Which kind makes sense to me.
I wish I could find where I saw that post but I can't at the moment. Anyhow it convinced to not go in that direction.

As an owner of an HT product I too have had "conversations" with JC, in fact, I've had the one you just mentioned. From what I observe when piercing thicker material, the key is to have the material blow out in as close to a concentric circle from the center as possible. If material starts blowing out on one side or the back (as in a ramp), it tends to cause uneven erosion of both the shield and the nozzle, prematurely shortening the life of your consumables.

phillc54 wrote: I recall reading something where in that situation you could do a first run where you just do the pierces then clean the sheet then do a second run for the cuts.

This is exactly where I finally arrived after struggling for months with many failed holes and burned out consumables.

Anytime I have to burn holes in anything 1/2" or thicker, I first don't even bother to try nesting my parts. I generate one part then immediately make a copy of the file. I then go into the first one and strip out every thing except for the actual pierce. I.e. Move to center of the hole, touch-off, raise to "my" favorite pierce height (not Jim Colt's) then blast most of the material out of the hole for about 1.5 seconds. Clean off slag after torch-off then go back to zero X and Y.

In the second file, I simply comment out the pierce delay time and set the plunge rate to something quick. The fact that I haven't completely burned through on the initial pierce ensures that there is material for the arc to start then get established before the lead-in commences. Using this method, I can now burn a full sheet (4'x8') of 3/4" material (1 foot squares with 4 to 6 holes) without a single reject. I do one at a time then simply index in the X or Y direction to start the next part. It's slow, but much faster than grinding and reaming holes that aren't round enough to get a bolt through.

phillc54 wrote:

thefabricator03 wrote: Phill,

Please see attached for the Gcode for the 20mm base plates I was having problems with.

When I pierced the holes the molten metal created a line of metal that went over the outside contour of the part. When I was cutting the outside of the part at cut height the torch would hit the line of molten metal.

This is a bit crude but seemed to work ok in a sim so it should at least prove/disprove the theory. If you want to try it then you will need to:
Rename your existing plasmac_gcode.py link in your config directory for safekeeping.
Copy the attached file to your config directory.
You would need two tools/materials set up for this, the original would be used for piercing and a copy with the pierce delay minimised would be used for the cut.
If you have #<pierce-only> = 1 anywhere in your GCode file the GCode parser will output a piercing only file.
So you could have two original GCode files, one with the pierce command and one without or you could have one original and save a new file after the GCode parser has output it and us that directly.
If it goes tits up just delete plasmac_gcode.py and rename your link back to it original name.
I hope I have explained it well enough...

This would work for my needs, but I'm not yet on the LinuxCNC bus. However, thanks to the many advances of people such as yourself and TommyLight, I'm getting close to pulling the plug on UCCNC. It's just a shinier version of Mach3 with even less support.

Pierce only mode is now a supported feature of Plasmac and thefabricator03 has tested on a full sheet of 16 mm with perfect results.

Sounds like you are running out of excuses to give Plasmac a try. The support from this forum will be better than any commercial plasma controller manufacturer. Somebody is here to help from somewhere in the world 24/7.

Stefan thefabricator03 threw out an AUD $12k investment in another well known commercial controller and has never looked back.

I'm not sure about solving your piercing issue completely. On the little bit of 16mm I have cut with my Thermal Dynamics A120 it was not a problem (120 amps, 5.1mm pierce height, 0.7 second pierce delay). My machine does list cut settings @ 80 amps but if you follow a rule of thumb of 10 amps per mm of thickness, that is only half what you need.