Chris Radek wrote:
> On Tue, Jan 27, 2009 at 10:08:46PM -0600, Jon Elson wrote:
>   
>> I think this is absolutely the correct way, as the C axis will then 
>> always follow the path, even when manually moving!
>>     
>
> Unfortunately not, since manual motion is currently in joint mode.
>   
No, I was thinking of using axes 0 and 1 at the HAL level to compute the 
C angle.
>
>   
>> And, there's no need for adding C movement to the G-code.  When you 
>> program a single arc move in XY, the C move would have to be programmed 
>> in linear segments, greatly complicating the G-code.  
>>     
>
> This was discussed earlier in the thread - rotary axes can move
> during arcs and this gives exactly the needed motion.  The
> calculations for generating this gcode are simple.
>
>   
Right, a constant-rate C move at the same time as a G02 arc would stay 
in sync just fine, I wasn't thinking straight about arcs being exact 
circular movements only, but of course they are.
>> HAL has to be the 
>> way to do this.  You could always use an optional digital output to turn 
>> this mode off if desired.
>>     
>
> I still think gcode is the best way.  Jeff's ideas about
> auto-generating with a filter are interesting too.
>
>
>   
That makes more sense.  I doubt anybody would be running a vinyl cutter 
with tool length and radius compensation, so the filter could be pretty 
simple.  Since, as you say, the C axis can be programmed with a single C 
word that corresponds to each arc move, and add a separate block that 
sets the C axis before each G01 move, the resulting code would only be a 
slight increase.  I don't know if vinyl cutters usually have any 
pre-locating of the cutter before each cutting move, or just treat the 
cutter as a point.

Jon

------------------------------------------------------------------------------
This SF.net email is sponsored by:
SourcForge Community
SourceForge wants to tell your story.
http://p.sf.net/sfu/sf-spreadtheword
_______________________________________________
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to