Re: Re: EMC to be used with MasterCam or ProE [asimtec]

From: "John Delaney" <john.p.delaney@...>

> OK, well I'm really new to this but my gut reaction is to go looking
> for lines of G-code that are making it do that and delete / change
> them.
>
> Or is it possible to tweak your drawing and thus tweak the output?
> Just a thought.
>
> Or you could just write it off as a "personality trait" :-)

A long time ago, the EMC did have a bug called NAN -- Not A Number. This
would cause the affected axis to race off toward infinity at warp factor
four. I thought that Ian Wright had found the last of these for us but
then that might be what I get for "tinking" as my neighbors would say.

You say that the problem is intermittent but it always happens during the
execution of a g-code program. I'm wondering if you begin and end the
program so that it sets standard or known values to all of the modal
command state. As a simple example, if cutter diameter or length comp is
active when a program is aborted or errors out, it is not turned off and
will cause some strange behaviors when the program is restarted. There
is a sample program in most releases, skeleton.ngc, that Matt wrote to
illustrate beginning and ending codes.

You can find the areas within a g-code program that cause problems, if you
set debug in INI to the max (7fffffffff) and run the EMC from a terminal
rather than from an icon. You will be able to get fairly close to the
offending line by watching the interpreter read and task execution
blocks. The only thing to remember is that the interpreter numbers lines
starting from 1 rather than using any N numbers in the program itself.

Hope this helps

Ray