Newsgroups: comp.robotics
Path: brunix!sgiblab!swrinde!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: <Cn7H4o.By2@unix.portal.com>
Sender: news@unix.portal.com
Nntp-Posting-Host: jobe.shell.portal.com
Organization: Portal Communications Company
References: <PJG.94Mar18132910@parint.esl.com> 	<1994Mar21.171526.28204@Mr-Hyde.aoc.nrao.edu> <PJG.94Mar21134448@parint.esl.com>
Date: Fri, 25 Mar 1994 05:41:08 GMT
Lines: 24

pjg@parint.esl.com (Paul Gyugyi) writes:

>By linear, I mean that for a _fixed_ load, a small variation in the voltage
>causes a variation in the speed which is fairly linear (*).  You could also
>whack the thing with a hammer to cause it to stop, but that wouldn't
>show anything about linearity either :).  But the real question of the
>thread was why to use PWM instead of a piecewise-constant output with the
>same average value.  The reason is that you get less heating of the motor
>with a PWM signal, and so you can go to a higher torque/speed without 
>overheating the motor.

I believe the heating would be the same either way, if any difference.
Heat is a biproduct of power dissapation.

P = V^2 / R;  assuming R stays the same since the coefficient for 
thermal conductivity for copper is + not - , the copper will reach
a given temp. under a given load; R is relatively constant.  The V
is Vrms.  Which for PWM would be the same as direct linear DC.
6Vrms is 6Vrms.  I do not see any way of more power being used to 
supply a direct linear control (not that I think it is the best, PWM
is much superior). 

-Kip 
arioch@shell.portal.com
