Plasma CAM: "Lead In" option

[Pensee]

A "Lead in" option would be very nice, for some tools like Plasma cutter.

• a great thanks to the dev's team for all the amazing work.

[ghost]

See also previous discussion here: https://github.com/Smoothieware/Smoothieware/issues/841

I would say we can leverage the existing laseron / laseroff code from settings, to allow users to customise their firing sequence., because even on same firmware, the process is very much machine dependant:

Here's a video of my machine (now sold with the move sadly) https://www.youtube.com/watch?v=0ZMn9tS2_II in action

So the basic start stop of a plasma works like this:

No Torch Height Control + simple chinese inverter

1. G0 to start of the lead-in (Because a plasma start leaves some blowout and a large hole (up to 3x the normal kerf) you want to init the arc next to the cut, in some waste material)

Example: Start in center of hole, cut to radius, then complete circle

CAM challenge: Algorithm to determine where to lead-in from (circles are easy = center) but such that its outside any other workpiece (always in scrap)

1.1 Note; May include a Z move to position it at optimal pierce height to start the arc

2. Start the Arc: Here with Smoothie we tend to use a Switch Module with a custom M-Code http://smoothieware.org/switch or for Grbl we usually tie off the Spindle pin In the case of a cheap chinese inverter, this just flips a relay module, closing the wires that usually went to the handheld torch)

2.1 Once the arc is triggered, the HF pre-arc usually fires for less than a second or so, before the main arc starts. On the cheap machines (and since neither grbl or smoothie can talk to a THC yet to read the ARC OK signal) we just rely on a G4 dwell determined with a little trial and error (longer for thicker material) to allow the arc to pierce the metal

2.2 Once pierced (or sufficient time surpassed) usually machines without a drag torch, needs a Z move here, to bring the nozzle closer to the work (pierce height is around 5mm above surface with a 40A chinese plasma, and around 1mm for cutting)

3. Then G1 to complete the Lead In and start the cut

4. Follow the cut (existing laser inside / outside cam will do from here - plasma kerf == laser beam dia)

5. At end of cut, switch off the Arc (custom M-Code or Spindle off) followed by (if not a drag torch) a Z move back to either zClearance or next Pierce height

6. Repeat from 1

[ghost]

So in LaserOn (we'll rename the field in settings to Tool On or something @jorgerobles

we'll typically have

G0 Z5
M3
G4 P400
G0 Z1

And in ToolOff

M5
G0 Z5

or if its a machine with a floating Z / drag torch

M3
G4 P400

And in ToolOff

M5

Since CAM can grab that from Settings, we can manage that with user specified or machine profiles

Th only real dev work is the Lead-In (;

[ghost]

Sheetcam does some extra cool lead in and also corner overshoot moves to improve quality even more

http://www.sheetcam.com/features/plasma

[ghost]

http://torchmate.com/cad-cam-7-8-9/FAQs-Tool-Paths good read as well

[jorgerobles]

Should do I change the Laser On/Off label to Tool On/Off or also the variable name (gcodeLaserOn.. Off) ? if the later, just note it @tbfleming (I will replace also his files, of course)

[ghost]

Well, we already have more milling operations than laser (; So we are pretty well ahead of overusing the word laser.

[jorgerobles]

Sooo I went ahead anyways :D also transform gcodeToolOn to a TextArea, for any requirements to come

[ghost]

I also went ahead and added PlasmaInside and PlasmaOutside to the Operations Diagram

[jorgerobles]

Then also waterjet ;)

2016-12-18 15:36 GMT+01:00 Peter van der Walt notifications@github.com:

I also went ahead and added PlasmaInside and PlasmaOutside to the Operations Diagram

[image: plasma] https://cloud.githubusercontent.com/assets/7695323/21294197/1b7ddb48-c540-11e6-9edf-8f2cac076344.PNG

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/LaserWeb/LaserWeb4/issues/87#issuecomment-267824656, or mute the thread https://github.com/notifications/unsubscribe-auth/ABoIYKHSk0KuyVOGLHku7F6L5K6x5hvrks5rJUT9gaJpZM4LQGfu .

[ghost]

Lol! WaterJet and Plasma pretty much are exactly the same - pierce time etc included

[jorgerobles]

I guessed that :D I suppose that diagram will suffice.

2016-12-18 15:38 GMT+01:00 Peter van der Walt notifications@github.com:

Lol! WaterJet and Plasma pretty much are exactly the same - pierce time etc included

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/LaserWeb/LaserWeb4/issues/87#issuecomment-267824793, or mute the thread https://github.com/notifications/unsubscribe-auth/ABoIYGbN6t98Hp2hRb5c92RJrRftZfRtks5rJUV9gaJpZM4LQGfu .

[tbfleming]

From those links, it looks like line lead in is better than curve lead in. Line lead in is also easier to implement.

[ghost]

Agreed, line is fine. At the feedrate of consumer plasmas especially (the industrial 300a units run at insane ipm's and the curve makes for less vibration, but with a 40A hobbyist plasma you are running around 800mm/min in 3mm mild steel

On Dec 18, 2016 5:30 PM, "Todd Fleming" notifications@github.com wrote:

From those links, it looks like line lead in is better than curve lead in. Line lead in is also easier to implement.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/LaserWeb/LaserWeb4/issues/87#issuecomment-267827563, or mute the thread https://github.com/notifications/unsubscribe-auth/AHVr2xYGQ4YgTIfQ40UrsnuRGIz-IgXaks5rJVGBgaJpZM4LQGfu .

[Pensee]

keeping a eye on it ...

[ril3y]

Did this ever get implemented? In could not find plasma operations in current code base.

[cprezzi]

Depents on if there is someone willing to implement it. The former cam dev has left the project.

[Pensee]

If only I had the ability to help it, But I am still very interest by this option ...