ATC code works great in isolation, all kinds of problems in program

I've got my tool changer working great in isolation, but when I try to run a program that changes tools, all sorts of things go wrong.

I've attached the files. The REMAP subrouting, and the two subroutines it calls (lt and ut to load and unload tools).

Things that go wrong:

1. When I load a program, while it's scanning through the file, it tries to load the first tool. Or rather, it calls my loading code, but doesn't move the machine, and because my code waits for a tool-lock signal, it aborts.
2. I can continue from this. Sometimes, it starts the tool change procedure, moves over the tool pocket, and then just stops moving. My code reports no errors, and program execution doesn't stop. The machine simply doesn't move.

Any ideas what might be causing this, or how to handle it?

JetForMe wrote:

I've got my tool changer working great in isolation, but when I try to run a program that changes tools, all sorts of things go wrong.

I've attached the files. The REMAP subrouting, and the two subroutines it calls (lt and ut to load and unload tools).

Things that go wrong:

1. When I load a program, while it's scanning through the file, it tries to load the first tool. Or rather, it calls my loading code, but doesn't move the machine, and because my code waits for a tool-lock signal, it aborts.
2. I can continue from this. Sometimes, it starts the tool change procedure, moves over the tool pocket, and then just stops moving. My code reports no errors, and program execution doesn't stop. The machine simply doesn't move.

Any ideas what might be causing this, or how to handle it?

I don't see any files attached.

Is it a python remap by chance? It sounds like it is not checking whether it is in mill task or not.

bevins wrote:

I don't see any files attached.

Is it a python remap by chance? It sounds like it is not checking whether it is in mill task or not.

Oh, so sorry! They're attached now.

It is not a Python remap; it's Gcode. Something like that makes sense, but does it need to be done in Gcode as well?

EDIT: I found the discussion around this concept: linuxcnc.org/docs/html/remap/remap.html#...apped-code-execution

Thank you, I should be able to fix issue #1. Not sure about issue #2. But I had to restart the machine for other reasons, and that hasn't happened again.

JetForMe wrote:

bevins wrote:

I don't see any files attached.

Is it a python remap by chance? It sounds like it is not checking whether it is in mill task or not.

Oh, so sorry! They're attached now.

I dont see anything that screams wrong...... I am thinking when you fix your milltask issue that should fix it maybe. The interpreter sometimes just ignores your commands in remap, if everything is not as it should be. I have struggled with that at times.

/EDIT load your ini, remap and toplevel files.

Have you modified your EMCIO section of your machine.ini file to include:

"TOOL_CHANGE_POSITION = 0 0 0"

(change the 0 0 0 to wherever your home position is).

forum.linuxcnc.org/49-basic-configuratio...-move-to-home#188561

rootboy wrote:

Have you modified your EMCIO section of your machine.ini file to include:

"TOOL_CHANGE_POSITION = 0 0 0"

(change the 0 0 0 to wherever your home position is).

forum.linuxcnc.org/49-basic-configuratio...-move-to-home#188561

This shoulkdnt have anything to do with his remap situation. This simply moves the machine to the XYZ coords when toolchange is called.

You are correct.

From how I interpreted the problem, I took the problem to likely be a "Not at Home" issue based on this line: " it calls my loading code, but doesn't move the machine".

Looking at it again, the machine in question could easily be the tool changer itself rather than the mill.