andy pugh wrote: > > > A single-count deadband in the PID might help. > > I am not sure if this is a good idea, but I wonder if a low-pass > filter on the encoder position would help? > This was discussed some time ago, the real problem is a low-pass filter must cause delay, and that is bad in a servo loop.
A better way is to do velocity estimation based on time stamping arrival of the encoder pulses. Unfortunately, the interface that Bruce has doesn't have this feature. The coarser the encoder resolution, the more jumpy the velocity computed from sample-sample difference has to be. He does have 1.5 counts worth of deadband in the PID already, but that equals .015 mm or .0006", so he can't make it much bigger. Jon ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity and more. Splunk takes this data and makes sense of it. Business sense. IT sense. Common sense. http://p.sf.net/sfu/splunk-d2dcopy1 _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
