Stuart Stevenson wrote:

> 
> 
> Gentlemen,
>    Thanks - Dave and Jon - just the info I was looking for. It looks
> as if new amps will be cheaper-faster-better.
Cheaper ?  No.
Faster?  Quite likely.
Better?  If the Gettys amp is truly single phase, it is a 
disaster, and I think it would cause serious hair pulling to 
make it work with EMC.  Any time the outer servo loop has more 
bandwidth than the inner loops, you get into maddening problems 
with the outer loops overcompensating and causing oscillation.
With the bandwidth mismatch 100:1 the wrong way, I'd hate to
see what would happen.
>    Jon, is it possible to parallel two of your amps to double their capacity?
You are talking about the PWM servo amps?  Hmmm, maybe.  If you 
just ran the PWM and direction Opto's in series, the PWM 
controller should be able to drive them both.  Or, it could 
drive the new version in parallel, as they need less current
in the optos.  Either way, that would have the PWM synchronized, 
and they each have their own current limiting, so they'd sort of 
share at their own limit points.  I'd probably better try this 
here before I ship anything, but I believe that would work.

My PWM amps take a PWM signal as their input, and so you lose 
the velocity and torque loops of a true velocity servo amp.
If the encoder has sufficient resolution, it is not a problem.
If it is a lower resolution unit, then you will get step-wise 
motion at low speeds.  I don't think you've worked with my PWM 
controller, just the PPMC.  They are mostly similar, but the UPC 
is not modular like the PPMC.

I have thought about making a new product that combines the 
analog velocity loop front end of my original analog velocity 
amp with the sign-magnitude PWM back end of my latest PWM servo 
amps.  (That original velocity amp used synchronous antiphase 
PWM, and the output filters ran VERY hot!  It had a few other
problems as well.)  I've never gotten the tuning of the thing as 
good as I'd like, there are some mechanical oscillatory poles in 
the system around 20 Hz that I've never been able to eliminate. 
  If I damp the servo amplifier too much, then the EMC loops get 
unstable, if I damp less in the amplifier, then the mechanical 
resonances go wild.  The thing is too close to the edge, and 
I've never been able to quite tame it.  I probably just need to
turn the gain down, but then the static error goes up, but maybe 
that is still small enough to be ignored.  Anyway, that is just 
an idea, and I'd have to do serious testing before saying such a 
thing exists.

Jon

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
Register now and save $200. Hurry, offer ends at 11:59 p.m., 
Monday, April 7! Use priority code J8TLD2. 
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to