2012/12/21 Peter C. Wallace <p...@mesanet.com>:
>
> OK thats roughly 125 ma/uSec at 28VDC power so ~3A ripple at the default
> 20 KHz PWM so I would set the PWM rate to say 40 KHz to cut this in half
>
> I would also make sure that you have the pwmgen deadtime set to 0
> and the PID comps max output set to .95
>
> (assuming the 3pwmgen scale is the default +-1)
>

Thanks! Set them and it seemed better. Is there anything else I should do?

I would appreciate some advice on getting the asymmetry somewhat better.
This HalScope plot shows ferror and pid output for the Y axis motor,
when I jog it back and forth with 4 m/min (motor is installed in
machine):
http://picpaste.com/asimetrija-hYmwX1Z9.png

The thing is that yesterday I spent 5 hours to realign hall sensors
for that motor (that involved also physical changes to the pcb of that
sensor to extend its "range of movement"). The result was that by the
evening this motor, when installed in machine, had symmetrical pid
output, as I jogged it back and forth. So now I think that it does not
make much sense to try to realign them again as there will be
asymmetry again tomorrow or after Christmas or whatever. How can I
check that those hall sensors are really 120 degrees apart each from
other?


BTW what velocity should I set, when tuning pid loop? As fast as
possible to ensure that it is stable, when going at full speeds or
something very slow to make it stable, when there are very few encoder
pulses per servo period?

-- 
Viesturs

If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

------------------------------------------------------------------------------
LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial
Remotely access PCs and mobile devices and provide instant support
Improve your efficiency, and focus on delivering more value-add services
Discover what IT Professionals Know. Rescue delivers
http://p.sf.net/sfu/logmein_12329d2d
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to