Has anyone had macro programs exit "early" allowing the next NC cmd to run?

Good afternoon all,

I am working with: MACH 3, Macros, Ethernet Smooth Stepper, and Windows 10.

I am developing some rather complex macros used in a tool changer system. Within any of the macros, there will be fixture offset changes (from G54 to G50 Pxxx) for example, motions on any of the axes, etc. I have tested this system for a couple hundred times and on two occasions the machine "took off" in an unexpected motion. While "****" was happening, I was not able to fully comprehend why it was doing what it was before hitting RESET. Of course, when you do that, the message is cleared, fixture offsets are reset back to G54 etc.

In both cases, the closest I can determine is the macro that was running, terminated early which then allowed the NC program to proceed to the next statement and execute it. In my case, that is running another Macro. I have not been able to find a way for the macro to terminate early for this to happen. This is determined by the state of all the devices, meaning several things in the macro executed, but not all of them.

As mentioned, this will run perfectly the majority of the time so it "ALMOST" seems random. But frankly, can't think of anything that is random that can cause an M-Code macro to terminate early.

Any ideas, suggestions, Oh Yeahs, I had something similars????

Thanks in advance!

Chris D

If a macro does not finish in Mach3, I highly recommend restarting Mach3 before doing anything else. Otherwise, bad things can happen.
Unfortunately, there are a lot of little bugs in Mach3 that only show up 0.5% of the time.

Now, I've never seen a macro terminate early for no reason, so it may be an issue with your macro.
The only thing you can really do is test, test, and test some more until it's 100% robust.