John Kasunich wrote:
> Jon Elson wrote:
> 
>>Peter C. Wallace wrote:
>>
>>>If you are just driving a bare motor, try raising your sampling rate...
>>
>>Geez, the resonance seems to be about 40 Hz, my sampling rate is 
>>already 1 KHz, 25 X higher.  But, yes, I can go to at least 5 
>>KHz and see what happens.
>>
>>Jon
>>
> 
> 
> Do you have current feedback or a current loop?  (I doubt it.)
No.  This is an absolute bare-bones drive.  Measuring current
is fairly easy on a brush drive, and I've done that before. 
But, I don't even know how to measure the current on a 3-phase 
motor.  My brush solution was to put a current sense resistor in 
series with the output right after one half-bridge (before the 
output filter) and read it with an Analog Devices AD620 
instrumentation amp.  This works amazingly well reading 100 mV 
drop off a resistor while the common-mode is a 70 V 100 KHz 
square wave!  But, I really don't see how to do that and 
decipher four-quadrant motor current from the motor lines.  The 
problem then is you need to get the current reading into an 
equation, while EMC is already putting out a digital PWM signal.
I only have a 36 macrocell CPLD on the brushless servo amp board
and 4000-series CMOS on the brush model, so DSP computations are 
out of the question.
   If you
> control current, you more-or-less eliminate one pole from the system
> response.
Yes, maybe this whole PWM idea has a fundamental flaw, 
especially with larger motors.  But, now I've backed myself into 
a corner, building a whole product line around a possibly flawed 
concept.

Jon

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
Don't miss this year's exciting event. There's still time to save $100. 
Use priority code J8TL2D2. 
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