Hi Al,
I suppose there are really two answers to this question. (a). write  
the g-code yourself, tedious but gives you exactly what you want or:  
(b) use a cam program that generates arcs with cutter comp.

Your question gives me a nice lead-in to bring up a small problem  
within emc. Cutter comp,  in emc, was set up to go two ways, assume  
the tool table knows the diameter of the tool and uses that ... or  
declare the dia of the tool in the cam program and let it generate  
the offsets. The cam I use (synergy) takes the second approach. The  
programmers for synergy did this with the idea that the cutter comp  
within the motion control would accept small numbers in the tool  
table to represent deviations in the tool diameter. That is,to  
correct tolerances in the part or compensate for tool regrind one  
just entered the deviation in the tool table and were good to go.
However emc barfs on the second approach. Everything is fine until  
one enters something besides 0.0 in the dia for the tool table. Since  
I expect emc to be a very capable program, thanks to the brilliance  
and dedication of the programs, I'm surprised that it does not  
contain this feature.

My local approach (read fix) is to declare a zero diameter tool and  
then modify my tool table.

Maybe one of the developers can explain how easy or difficult this  
would be to fix/implement.

Thanks for listening.

Dave


On Jul 30, 2007, at 12:00 AM, Al Hooton wrote:

>       I'm getting going with emc2, after not having done nc/cnc work for a
> long time (like, almost 30 years ago, before software took me away  
> from
> machining as a career).  I've been messing with the G41/G42 tool comp
> stuff, and have the lead-in moves working correctly.  However, I've  
> got
> a question...
>
>       I have run into the problem I see mentioned every now and then in the
> archives, wherein the tool comp algorithm has problems with series  
> of G1
> moves that were generated by some higher-level application to  
> represent
> an arc.  This makes sense, since emc2 can't maintain tangential  
> contact
> with the material path at the vertices on the tool path.
>
>       I was surprised to run into this, as my expectation was that these  
> arcs
> would have been represented by equivalent G2/G3 arc moves instead of a
> series of G1 moves.  So, I started digging into the code behind every
> open-source package I can find, including OOdraw, inkscape, qCad,
> pstoedit, hp2xx, dxf2gcode, codeg, and probably some others I'm not
> remembering right now.  I have also reviewed every vector-based file
> format ever invented.  I consistently run into the same problem as  
> I try
> to build a software tool chain: somewhere between the cad/drawing
> package and the g-code generator, one of the apps is converting  
> arcs or
> splines to series of lines.
>
>       In my most recent attempts, the source drawing (OOdraw or inkscape,
> exported to pdf, svg or dxf), has every corner radiused more than  
> enough
> to allow for the tool to make the cut.  Unfortunately, I can't get  
> those
> radiused corners all the way down to g-code as arcs.
>
>       I am determined not to use windows, and also to avoid paying for
> software if I can manage it.  More as an experiment than anything  
> else,
> but I'm an old-school open-source guy and I just figured there  
> would be
> kickin' solutions for all this stuff in open-source space at this  
> point.
>
>       What am I missing?  How can I get a high-level 2d or 3d drawing  
> all the
> way down to emc2 in g-code such that arcs are represented as G2/G3  
> moves
> instead of series of G1 moves?
>
>       Thanks!
>
> -Al
>
>
>
>
> ---------------------------------------------------------------------- 
> ---
> 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


-------------------------------------------------------------------------
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