Newsgroups: comp.robotics
Path: brunix!news.Brown.EDU!agate!dog.ee.lbl.gov!ihnp4.ucsd.edu!pacbell.com!decwrl!portal.com!portal!arioch
From: arioch@shell.portal.com (Kip J Mussatt)
Subject: Re: PWM DC motor control
Message-ID: <Cn6uy3.Dtt@unix.portal.com>
Sender: news@unix.portal.com
Nntp-Posting-Host: jobe.shell.portal.com
Organization: Portal Communications Company
References: <Mar18.181352.31507@acs.ucalgary.ca>
Date: Thu, 24 Mar 1994 21:42:01 GMT
Lines: 30

morrow@cns8.cns.ucalgary.ca (Bill Morrow) writes:

>jpeskin@bear.WPI.EDU (Jonah Peskin) writes:

>]    I'm looking for ways to control the speed of a few DC motors I wanted
>]to use to drive a robot.  The only thing that I am aware of that will do this
>]is to use a pulse wideth modulated oscillating signal to drive some power
>]transistors which drive the motors.  The idea being that a lower duty cycle
>]signal will rsult in slower motor rotation.  

>Is anyone trying to actually vary the driving voltage to the motor,
>using some feedback from shaft encoders? You know, just like they 
>taught you in that control systems course. It seems to me that this
>type of control system would be more efficient, responsive and cover
>a wider range of motor speed.

Hmm, my controls control class began with the simple "change the voltage"
hinted at varying the pulse width which we didn't cover because of instructor
mental limitations.  In a later class, I learned the thing to do is use PWM.
This helps control motor speed esp. when friction is involved since uK < uS.
In modern control systems it is much easier for a computer to vary an output 
pulse which controls a driving circuit than to use the same amount of data
bits to do a DtoA and then send the analog signal to the same driving circuit.
This second method is more costly and doesn't have the precision the first
method is capable of.

Just my $.02

-Kip
arioch@shell.portal.com
