Richard Arthur wrote:
> A couple more questions with reference to:
>
> http://imagebin.org/73098
>
> Looks to me like harmonics in the constant speed section; any 
> suggestions as to its source?
> There are spikes at the change constant speed/acceleration and vice 
> versa. I wouldn't expect to see that in a smooth transition, any comments?
>
> My setup:
> Brushed DC motors with tacho feedback to servo amps (+-10v control
What are you COMPLAINING about????  The worst-case peak-peak error is in the
neighborhood of 300 uInch, or 0.0003".  Unless you are diamond-turning 
telescope
mirrors, this is probably FAR smaller than any other error in your machine.
Since these motors have discrete windings, commutator segments and maybe
belt drives, there will be perturbations as they turn.  Also, you have 
the inevitable
double quatization noise of the encoder.

As for your spikes, you should plot velocity as well as position.  There 
is an acceleration
"jerk" at the transition from rest to constant accel, and then another 
at the transition from
accel to constant velocity.  (There is an experimental S-curve traj 
planner, but I don't think
it was ever made mainstream.)  I always view velocity (called "delta" on 
my driver, but you
may have to manufacture such a signal with a hal component) with pid 
error when tuning.

Jon

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to