Dynomotion

Group: DynoMotion Message: 8371 From: Hugh Sontag Date: 9/17/2013
Subject: Tuning
Hi Tom,

I did some tuning of the Y axis on our CNC router with brushless DC motors today.

The response is pretty good, but I'm not clear on how I could improve things with the IIR filters. There is some audible noise during a move, which I suspect is due to the oscillatory drive to the motor. The same oscillatory behavior can be seen in the position data.

Also, I can't make much out of the Bode plot. It may be that the data isn't really valid. I notice when I turn smoothing off, I have a field of data points that don't have a lot of correlation.

I've put a copy of the response to a Move of 10000 counts (I have 2048 encoder counts per motor revolution) and a Bode response plot at the URL


--> Could you take a look at the Move response and make any comments or recommendations regarding the tuning? Any IIR suggestions would also be appreciated - could you also explain what criteria you used to decide on an IIR strategy?

--> Could you comment on the Bode plot? The magnitude is always below 0 dB, hard to believe.

Thanks,
Hugh
Group: DynoMotion Message: 8373 From: Tom Kerekes Date: 9/18/2013
Subject: Re: Tuning
Hi Hugh,

As you point out your servo is unstable while moving with an oscillation at around 60Hz.  This often happens if you are relying on friction to provide damping (when moving there can be less friction).  This is a hard phenomena to see because it is a non-linear behavior (sometimes there is friction and sometimes there isn't).  A Bode Plot is based on a linear system that behaves the same regardless of speed or magnitude of signals.   

When making a Bode plot it is important to have the right stimulus magnitude and frequency.  Too little encoder movement and there will be no useful information, too much and the amplifier will be in a non-linear mode (saturation) not representative of normal servo operation.  So include a Time Domain plot so we can also have an idea if the Bode Plot should be meaningful.

You included the raw data files which is good because we can then plot and zoom as we wish to view the data.  But it doesn't include any of your settings.  You should include that also (Config, Step, and Filter Screens).  Most likely you will need to reduce gains to be stable and/or increase D Gain or add a Lead/Lag Filter to add phase margin.
Regards
TK