Greetings all;

My original config for this had hal connections in all 4 axis like this:
loadrt  [KINS]KINEMATICS
loadrt  [EMCMOT]EMCMOT servo yadda + numdio=12
later...
net     z-pos-cmd       joint.2.motor-pos-cmd
net     z-pos-cmd       pid_z.command
net     z-vel-cmd       joint.2.vel-cmd
net     z-vel-cmd       pid_z.command-deriv

But somehow, this joint.2.vel-cmd has a huge high vel spike as it 
approaches zero velocity, jumping to the same offset in the other 
polarity, then decays back to where it would intersect a straight line 
continuation of the original trace, at which point it then continues 
upwards as if its a position command. And it does it at both polarities 
of a reversal. This spike is way faster than the z motor can usually 
move, and results in a z following error at the bottom of a g33.1 move.  
This is despite a limit3.maxa and maxv setting in series with pid_s that 
results in a quite leasurely reversal at the bottom of the stroke. 
Everything else is supposed to just follow the spindle, and its not.
In the  halscope display, the encoder output is following this restricted 
reversal quite nicely. There is some pid_s.error showing during these 
velocity changes of the spindle, but its a gentle curve before and after 
this velocity spike.

So I have to figure out where the wrong signal is getting back to motion 
from because its all waaay out of time visually now that I have the near 
problem solved.

Has anyone else encountered this, or is it some sort of a bug in motion?

The man page for pid says the deriv will be calculated if its not 
connected, so would I be better off it is disconnected?

deadband is set to 0.0001" in that pid, but IMO its being fed bad data by 
motion.

Ideas? I sure could use some.  Thanks.


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to