New feature suggestion - what do you guys think ?

Special attention for Phil, Snowy and Rod (and anyone else who may be interested).

Qtplasmac looks downright awesome, but here's another idea for you guys to ponder.

Regarding torch touch off. Yes you can probe at full Z axis rapid speed, providing your touch off switch has enough over travel to allow for the deceleration distance, BUT..............there's the issue of the nice expensive torch slamming down at rapid speed into the sheet. Wouldn't be very good for the torch after a few hundred or thousand touch offs at rapid speed.

For quite some time I've had an idea/concept I call my "Feather Touch Off". Imagine some sort of vertically actuated micro switch mechanism with built in overrun. The force required to operate just a microswitch is very small and would cause hardly any deflection in even very thin sheet.
The beauty of this concept is that it eliminates the need for both a floating Z axis AND an ohmic sensor, a considerable cost saving. Of course a con is that you loose the ability to use the typical float switch as a breakaway trip input, but I'd prefer a proper breakaway in any case for pure side loadings where a float switch would not be activated.

This feather touch off mechanism would require an activation output that say operates an electro-magnet or a tiny pneumatic cylinder to bring it down to touch off position.
This touch off position would have XYZ offsets from the torch nozzle position, so these would have to be programmed in the Qtplasmac setup.
After the feather touch off, once the switch has retracted, the torch rapids over to the touch off location and pierce height.
On production cuts and thin dirty metal (bad for ohmic) this could make a significant difference to cut time, as every single touch off could be done at Z rapid speed.

Any opinions, am I missing anything, is the idea completely stupid ?

Keith

Les from sheetcam posted some shots of his plasma cutter recently and he uses something similar.... trying to find the clip sure I saw it in the last 2 weeks

Swiftcut has also published some videos of a probing "foot" as you describe.

I think most of the functionality as already there. The probe enable signal could fire your probe down and the offsets could be managed using the multiple tool/spindle functionality if the probe was another tool that had the x,y,z offsets programmed into it. So by selecting that tool (eg. like an engraver), it would move to the probe offsets. Seelcting the plasma tool, would bring the torch back to that position. You also get a bonus as the spindle relay for that tools spindle is also available due to Linuxcnc's muti spindle support. All that Plasmac has to do is to change tools before firing the torch.

Thanks guys,

well if Les and Swiftcut have already implemented this idea I guess it's not completely daft LOL. I was worried I'd missed something rather obvious.

Rob, definitely post a link to the shots of Les plasma cutter if you find them, would like to see it. I'll search in the meantime.

Rod, that's great info, I'll archive that for later.

Keith

Cheers Rob,

I also managed to find the Swiftcut probe foot video in their intro video. It's at 1:38 and this video

Keith

It would be reasonably trivial to add some code to get something like that operational. I would have two modes of operation:
1. Apply the X/Y offsets while probing if you have enough Z room to reach the offsets before hitting the material
2. Apply the X/Y offsets then begin. probing if you don't have enough Z room to reach the offsets before hitting the material

After a succesful probe, while moving to pierce height if would clear the offsets simultaneously but would of course wait for both the offsets to be cleared and pierce height to be reached before firing the torch.

The default mode could be set in the ini file (along with the offsets) and/or you could also have a User Button toggle the mode.

If your probe tip was at a distance below the torch tip greater than pierce height and you used mode 1 then you would use marginally less time probing as a byproduct.

If you want to do some guinea pigging then i'd be happy to do some sample code.

Wow, cheers Phill that's awesome.

I'm in the process of learning / installing / test running Linuxcnc and the new electronics. Have just tested my table with the basic install and am now modifying my Z axis home switch setup to suit Qtplasmac.

Once I'm up and running with Qtplasmac I'll try and rig up some separate probe foot and I'll come back to this thread to start guinea pigging.

I'm afraid I'm a bit confused and not sure what you mean when you say, "have enough room to reach the offsets before hitting the material".

Keith

I was thinking of just jamming it into the existing procedure...

1. G0 to next pierce point
2. M3 $0 S1 start cut procedure
3. extend probe, begin probe down, move to probe offset (simultaneously)
4. find surface, set zero
5. retract probe, move to pierce height, remove probe offset (simultaneously)
6. when pierce height and zero probe offset achieved turn torch on etc. etc.

#3 may be tricky if cutting something that is close to Z max in that the probe offset may not be reached before the probe is at material height. This could put the probe tip into a void.
To overcome that we have a halpin that could be set to tell it to split #3 into two parts:
a. extend probe, move to probe offset
b. when probe offset reached begin probe down

Of course we could do away with #3 and the halpin completely and just do 1, 2, 3a, 3b, 4, 5, 6 ...
It is just an option that may speed things up a bit and may be handy for folk that only cut sheet.

For those in a production setting, anything that speeds up probing has to help. Just looking at the stats in my version of plasmac:
35 min cutting
53 min rapids
72 min probing

I guess it show how important probing and rapids are to cycle times..