GCodestopped when encountering first line of # statement

[Kyle]

Hi guys (or gals if there are ladies in here),

I'm Kyle from Singapore. Just started making PCBs on my A2ZCNC/Sherline mill designed by a friend of mine. He was also the one whom introduced me to Autoleveller when I started having problems with uneven copper boards.

My first test with the test version of the autoleveller didn't go well. The software is simple to use. The codes generated looks good. But Mach3 refused to carry on after probing the 3rd time, when it encountered the assignment of the z tip position to the variable #500 (#500=#2002).

(begin initial probe and set Z to 0)
G0 X0 Y0 Z0
G31 Z-10 F100
G92 Z0
G0 Z2
G31 Z-1 F50
G92 Z0
G0 Z2
G0 X0 Y0
G31 Z-1 F100
#500=#2002

The message at Mach3 status bar: "begin initial probe and set Z to 0".

I'm using Mach3 (Version R3.043.062) with a USB SmoothStepper and Gecko G540. Eagle was used with PCB GCODE to generate the NC file. The screenset used (if it matters) is the BigTex Machblue screen.

I believe it has to do with my settings as my friend managed to get his working in the first attempt.

Anyone can help?

Thank you.
Kyle

[daedelus]

Not sure about this one. Its as if the memory space Mach3 reserves for variables like #500 etc. is locked somehow. Not sure how or why this would be. I am clutching at straws.

There may be some setting in Mach3 that someone can point you to?

[joeaverage]

Hi,
Kyle and I have been conducting a conversation on the Mach forum about this.

From that conversation we have established that the probing is working but certain
operations using pound variables, #500 #2002 etc are not working.

There are some specific tools that can be used and are working thru the various ops trying to
establish the cause.

As Al works perfectly on my machine I'm increasingly coming to the opinion that Kyle's Mach3
installation might be corrupt.

To my knowledge there are not any settings that affect the operations involving pound variables.

Will report back in due course.

Craig

[Country_Bubba]

Hey guys, I am NOT a Mach user so take what I have to say with a grain of salt.
But I often see on the zone people recommending one of the older versions of Mach over the latest because of bugs. Could this be a source of the problem in this case???
I don't know, but just offering some food for thought.

[daedelus]

In addition to what has already been said, another option is...
The Raw Probe File (RPF) options in the member versions of AL and AE.

Instead of using # parameters, probe values are logged to an external file which is then fed back into AL. This method removes the need for parameters so I presume you would not have the same problems this way. See: https://www.youtube.com/watch?v=ZbfS92Z ... e=youtu.be

Just an option. I am not trying to push you into becoming a member.