Another plasma component...

@JT,

In your hal file you can specify your limit switches. For your pin's open your linuxcnc application and open up halmeter to see what pin's are availble for your interface card. So look at your card.. and i/o names... You can write them down for practice.

net both-home-x1 <= lcec.0.digital_input.din-2 #(lcec.blabla is my sensor pin on left machine side x axis)
net both-home-x2 <= lcec.0.digital_input.din-3 #(lcec.blabla is my sensor pin on right machine side x axis)
net both-home-x1 => joint.0.home-sw-in
net both-home-x2 => joint.1.home-sw-in

This above is one way. But you the same will be :

net yourname => joint.0.home-sw-in => lcec.0.digital_input.din-2
Or without => sign's, use tab's instead :
net yourname joint.0.home-sw-in lcec.0.digital_input.din-2 ( in your case it could be mesa.0.digital_input.din-2 )

Above is all the same connnected, but you can connect in different way's.

JTknives wrote:

A quick question. Where and how are you guys calling out your axis limit switches. I have my float switch working but my XYZ Reed switches and not. When I look in the machine.hal I see reference to limit switches but no place to spec a pin as thy are labeled as joint. Also how do I use a single limit switch as a homing switch and limit switch. Thanks guys - JTK

JTK, its all described in the doc believe it or not.
linuxcnc.org/docs/devel/html/config/ini-homing.html
Just look in halshow and look for the pin names for each joint
With V2.8 homing is per joint.
For shared limit/homing You probably will need

I know which pins are which but when I went to master branch it set me limit callouts diffrent. This is what I have right now

My limit switches are
Are you saying I just do it like this?

net hm2_7i76e.0.7i76.0.0.input-04 => joint.0.home-sw-in

I would do this, net needs a name after. It was'nt me designing the linuxcnc hal side software....... :

net mypreciousconnection_or_your_name_you_want_to => joint.0.home-sw-in => hm2_7i76e.0.7i76.0.0.input-04

Or the same in less words :

Forget Positive or negative if random x axis limit is not working. So your example joint.0.home-sw-in is a good choice !!

Why can’t you assign a pin stright to an joint. Seams like having to have an Intermediary variable name to transfer data wasts code.

@Jt,

It would be nice if that was possible. But hal is constructed, designed this way.
So that's the reason i say in previous post : It was'nt me designing the linuxcnc hal side software......

In fact the core code of hall is limited. Try to connect one input to 2 outputs in one line. Then you need a component.
Hal could be designed more core like this :

void{input1 = output2; input1 = output3;}

In the source code we are able to make this progression's. In the source code we can dig into hal and make some extra argument's possible. That it is not possible at this moment if you do a git clone from source, there is no repro that defines a pure c hal environment.

This is only because hal read in is written to read a certain level and combination of words, net command's, exepts => sign's etc.
But smart student's can give it their own way. It's unlimited.

Also I had another question which is about the plasmac. How is the Z axis set up. At first I had my z set up so when jogged it would go down when you pushed the negitive button and up with positive. In the milling world I would set g54 as my as my zero from home position on X and Y and then Z is the difference from tool touch off location and top of part. But on a plasma how does the Z work because plasmac controlls all the z moves. Reasion I ask is because I was getting weird moves on my dry run and had to set the probe height to line 2” or it would detect ohmic when moving down befor it was ready to probe. Here is a short vid showing the dry run.

You are not really connecting a joint, you are connecting a pin to a pin via a signal

net x-home hm2_7i76e.0.7i76.0.0.input-04 => joint.0.home-sw-in

Signals (like x-home) are actually memory locations that are used to pass information
from a source pin ( hm2_7i76e.0.7i76.0.0.input-04 in this case ) to one or many
destination pins (just joint.0.home-sw-in in this case)
.

There is a (deprecated) way to directly link pin to pin: linkpp
linkpp still uses a signal, it just uses the first pin name as the signal name

linkpp joint.0.home-sw-in hm2_7i76e.0.7i76.0.0.input-04

Would create a signal joint.0.home-sw-in, linkin the pins joint.0.home-sw-in and hm2_7i76e.0.7i76.0.0.input-04. I suspect linkpp is deprecated because of the possible confusion of signals and pins

@Jt,

In plasma world you don't work with G54. You can use the benefit's of G10 if you want to have a plate switch position in the gui. Then simply press the goto home button for plate switch position.
G10 L2 P1 X0 Y0 Z0 (clear offsets for X,Y & Z axes in coordinate system 1)

For torch height zero point. The torch zero point is 3 to 4mm above plate or product. No need to use G54.
In sheetcam you can define it several way's. Easyes't way is the way above. Your probe or thc will do the trick for you.

In fact if your probe is on, you can zero at 50mm above the plate if you want. If you set your probe trigger to -15mm it will probe to that distance. All the logic is implemented in a good thc. The next probe will be done at a your given x,y distance and probing will start at travelheigh. So not at 50mm but at 10mm if 10mm is active in your screen. So the logic is inside the thc.

linkpp or linksp i never used, does it work until today pcw? It's a opposite connection request. Very confusing, just as you argumented.

linkpp still works (on 2.8) but the hal file parser complains about it