On Sat, Oct 06, 2007 at 09:55:21AM -0500, Stuart Stevenson wrote:

>    I need to implement 5 axis tool length compensation. The
> calculation is simple. Would the implementation be better in EMC or
> HAL?

Tool length compensation is currently in the canon level, which is
between the interpreter and the rest of EMC proper.  There are
currently X and Z offsets and these are simply added to the number
the interpreter asks for when it parses a line of gcode.  For
example if the interpreter wants a traverse to Z=1 and the Z tool
offset is 1, the canon level translates this to Z=2 and sends it on
to the rest of EMC.

Adding length compensation in this style for the rest of the axes is
not too hard, only tedious.  It touches many levels of the code from
the tool table (whose format would have to change again) through the
guis (some of which show these values when you load a tool).

However the big question in my mind is whether this is what 5 axis
tool length compensation even means.  Is it this straightforward?  In
a 3 axis mill or a 2 axis lathe, the compensated axes are
independent/orthogonal so it's easy to see how to offset them (you
only do the addition I described above.)

Is it like this for the general 5 axis case?  Is there even a general
5 axis case since 5 axis machines come in so many configurations?

I don't have an understanding of how, in general, a 5 axis machine is
controlled/programmed at the gcode level.  You've told me before
that short G1 moves are used in inverse time mode, which makes me
think the control really doesn't have to know much about the machine;
that specific knowledge about its configuration is left to the
programmer and/or CAM.  How does length compensation fit into this
picture?

Chris

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to