M6 Remap - Ignores M65

I'm having a problem with my NGC code in an M6 remap. The NGC pretty much consistently ignores any M65 command - leaving the signal on perpetually (until I override from the MDI ... which consistently works with an M65 Pxx command).

Any idea what to check?

It does seem strange that the 'splash program' displays the GCode and a highlighted code line (in Axis) when I command M6 Txx, but not sure of 'normal' behavior for that, as on our other mill (which doesn't have a toolchanger), it's been a long time since I've seen that splash program...

Any help is appreciated!

I can not help but i had a similar problem while seting up the tool changer on the Elumatec, namely one of the outputs kept stayng on even after e-stop was active, so only m65 p0 or restarting Linuxcnc it would turn off. I just was not pleased with that so i found an emco comp that ArcEye made and modified that to my needs. It works perfecrly. No remaps no python, just a comp and hal. It does need feedback from tool changer, in my case 4 pins, and they are easily modified in comp to suit your needs.

I am a bit surprised by this, as this Vismach config uses M65, and worked when I wrote it:

github.com/LinuxCNC/linuxcnc/blob/master...hange/toolchange.ngc

Can you run that config and see if the behaviour is the same as you are seeing with your real machine? (the sim won't try to operate any hardware).

Finally had a chance to get back and test it this morning. Just running the SIM, it seemed fine. Watched the hal pin for motion-digital-out-00 and it turned on and off during the toolchange, so I assume M65 is working properly in that code.

2 things:
1) Like I mentioned, M65 works in the MDI in my 'real' config ... but not (reliably) in the NGC code. What could be 'overriding' it? I know the NGC code is executing, it hits the abort() lines (typically on the very next line from the M65 command just fine ... the output just never turns off.

2) Even on the SIM, it displays the splash "LINUXCNC" engraving program ... and runs through the lines as if it's running that program even though it's actually running the NGC code (and seems to do that just fine). This is also what it does in my 'real' config. Is this a red herring and just expected behavior because the current NGC program hasn't been 'loaded' into Axis since it's called through a remap? Or does this point to a problem?

Thank you!

Thank you Tommy.
I may have to resort to that if I can't clear this up, but would really like to avoid having to write/edit a COMP right now ... especially since I have the logic of the NGC / PLC combo nearly working right now.

Right now, I'm also struggling with the machine E-Stopping right after orientation, but that is coming directly from the machine, and seems to be a problem with the machine circuitry expecting something that I haven't figured out yet...

Also ... it seems to be ignoring M64 P0 randomly right now as well. All works fine from MDI, but not from the atc.ngc