On Thu, 9 Jul 2009, Leslie Newell wrote:
> Date: Thu, 09 Jul 2009 00:33:00 +0100
> From: Leslie Newell <les.new...@fastmail.co.uk>
> Reply-To: "Enhanced Machine Controller (EMC)"
> <emc-users@lists.sourceforge.net>
> To: "Enhanced Machine Controller (EMC)" <emc-users@lists.sourceforge.net>
> Subject: Re: [Emc-users] Servo tuning
>
> Hi Peter,
>
> My point is that the I term is introducing the error in the first place.
> As you can see from the second plot, without I the response is pretty good.
Two gueses: 1. Your first accel error spike sets the integrated error to
some large value, and the time constant is so slow you get a very slow
recovery. Not sure why decel didnt drive the integral error the other way
Basically I wouldn't try to set "I" until you are doing motions that the
hardware can follow, otherwise integral windup will bite you. Also your
integral time constant looks too slow (but maybe its ok for such small error)
When you get accel set to a reasonable value, you might try increasing I
>
> I have been wondering if my problems are hostmot2 related but I can't
> think of any way that they can be. The encoder appears to be reading
> correctly and the output is simply PWM - I can't see how that could
> affect the tuning in the way I am experiencing.
HostMot2 is not related to PID other than feeding it position and velocity
and generating PWM. I'm pretty sure this is PID tuning issue\.
Peter Wallace
Mesa Electronics
(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.
------------------------------------------------------------------------------
Enter the BlackBerry Developer Challenge
This is your chance to win up to $100,000 in prizes! For a limited time,
vendors submitting new applications to BlackBerry App World(TM) will have
the opportunity to enter the BlackBerry Developer Challenge. See full prize
details at: http://p.sf.net/sfu/Challenge
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
