Re: 4th axis working

[Dan Falck <dfalck-at-sprynet.com>]

Would it be possible to mix and match parts of EMC?  Say that I only wanted 
to use the new rs274ngc code and liked the stability of the August 2000 
version.  Not being a developer, (or coder of anything for that matter) I 
am just curious.

Dan

At 11:08 AM 2/17/01 -0500, you wrote:


>On Fri, 16 Feb 2001, Dan wrote:
><S>
> > I ran into a bug having to do with tkemc.  When I first open tkemc, it
> > comes up defaulting to G21 metric mode.  If I go into MDI right away and
> > give it a G20 command and then try to issue another command, it locks 
> up my
> > computer hard.  If I start tkemc and go right into manual mode and try to
> > move the machine right away it locks up too.  So, I now open it, get it 
> out
> > of estop, turn on the machine, THEN press the MDI and Manual buttons
> > alternatively a few times and things will work fine.
> > Another bug in the new version is the inability to restart a program in
> > Automatic mode, without having to open the same program again. So, if you
> > are running a lot of parts and have to do multiple cycles, you will 
> have to
> > open the part program every time you want to start the cycle again.
>
>I think that I've been plagued by this bug for months now.  Recent posts
>suggested that it might be related to caching in the computer that runs the
>EMC rather than the code itself.
>
>When my system is going to crash is is usually near the start of some kind
>of motion.  Home or program start or a few lines into a program.  Home is
>instant.  I have been watching runs with the full debug set in the ini
>file for a while.  For crashes when I home out, it's always on the first
>axis and the terminal doesn't even report the NML sent. (I know that some
>terminal info is cached)  Some crashes seem to be associated with the
>burst of NML sent when the interpreter gets its run command.  Sometimes
>before any NML shows and sometimes after the 120 or so lines of code are
>sent in the initial burst.
>
>I'll try the idea of switching between modes a few times before any motion
>and see what happens.
>
>Does anyone have a further report on the effects of shutting down
>caching.
>
>Ray