Eric H. Johnson wrote:
> Sebastian,
>
> The servo period isn't much of a problem, with a 1.6 Ghz dual core processor
> I was able to get the servo period down to 250 to 300 micro-seconds before I
> started getting overruns. The problem I have not figured out yet is how to
> outfox the kinematics, since in 200 micro-seconds, the laser can go from
> full off to full on, however the kinematics will limit it to the feed rate.
>
> For the time being I am using really small values of Z, and then use a large
> scale value on the output. I have also been playing with using an angular
> axis but that has a few issues as well.
So there are two problems we're dealing with there, right? One is the
mysterious following errors, and the other is how best to drive the lazzzor.
I think I agree with Jim Coleman - using an axis position to give the
laser its power seems a bit strange. You want it accessible to the
g-code, Jim suggested spindle speed, but (if i understood you correctly)
you said you didn't like the abrupt changes in duty cycle that come from
using motion.spindle-speed-out directly.
Maybe you could use spindle-speed-out as an input to a properly tuned
limit3 component, and use the limit3 output as the laser power?
That would give you full control over both "rate of change" (the
"velocity" at which the pwm duty cycle changes) and "rate of change of
rate of change" ("acceleration" at which the duty cycle changes),
independent of kinematics. Seems simpler.
<http://www.linuxcnc.org/docs/2.2/html/man/man9/limit3.9.html>
But, that said, let's debug one thing at a time. ;-)
--
Sebastian Kuzminsky
"Okay, people. Now is the time to start discussing the rules of war for
autonomous robots. Now, when it's still theoretical." -- Bruce Schneier
------------------------------------------------------------------------------
SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
The future of the web can't happen without you. Join us at MIX09 to help
pave the way to the Next Web now. Learn more and register at
http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
