Michael Haberler wrote:
> motion has an issue with ferror calculation: 
>
> during a motion cycle, ferror is computed before the new commanded position 
> is calculated, and based on the difference of feedback (encoder position) and 
> the *last* commanded position. This means that ferror - which should track a 
> PID's error signal quite closely - actually deviates substantially from 
> pid.error and may also trip following errors when in fact there isnt one. I 
> understand this to be an issue with torque-mode loops.
>
> This patch makes the behavior of motion configurable by the new 
> motion.ferror-mode pin.
>
> By default (unconnected or set to 0), behavior is as before.
> If set to 1, ferror is computed only after the new commanded position is 
> known. This also means that the decision to raise a following error is based 
> on the new commanded position.
>
> for review, the patch is here: 
> http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=commit;h=dafe9eab55f209e56d08276efd65f62684b77f57
>
> I will commit this into master; I leave it open as to v2.5_branch, feedback 
> welcome.
>
>   
WOW, gotta try this!  As I may have mentioned before, this sounds like 
the exact situation
I have seen before with my motion interfaces.

Jon


------------------------------------------------------------------------------
Master Java SE, Java EE, Eclipse, Spring, Hibernate, JavaScript, jQuery
and much more. Keep your Java skills current with LearnJavaNow -
200+ hours of step-by-step video tutorials by Java experts.
SALE $49.99 this month only -- learn more at:
http://p.sf.net/sfu/learnmore_122612 
_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to