Newsgroups: comp.robotics
Path: brunix!cat.cis.Brown.EDU!agate!howland.reston.ans.net!gatech!panther.Gsu.EDU!news-feed-1.peachnet.edu!umn.edu!gold!roger034
From: roger034@gold.tc.umn.edu (Brynn Rogers)
Subject: Re: Selecting DC motors for a differential drive system
Message-ID: <CrDJ4F.LJ6@news.cis.umn.edu>
Sender: news@news.cis.umn.edu (Usenet News Administration)
Nntp-Posting-Host: gold1.tc.umn.edu
Organization: University of Minnesota
References: <2ti37h$cq5@info-server.bbn.com>
Date: Tue, 14 Jun 1994 06:23:25 GMT
Lines: 55

In article <2ti37h$cq5@info-server.bbn.com>, Deb K. Roy <dkroy@bbn.com> wrote:
>
>I am planning to build a mobile robot with a differential drive
>system.  I have sent out for catalogs from a few surplus DC motor
>suppliers, but I have no idea what torque I need.  The robot will
>weigh 50lbs, have 5 inch diameter wheels and run on carpet indoors.
>How can I estimate the torque requirements of the motor?
>
>I recently found a cheap ($30) motorized car that children can ride, and took
>out the drive mechanism.  It seems to be powerful enough and the
>right speed.  The gears are plastic and I'm sure it's a pretty low grade
>DC motor, but it was built to carry heavy loads.  What are the disadvantages
>of using toy parts like this?  Do the gears wear down quickly?  Are
>cheaper motors more difficult to control?  Are they going to be less
>efficient?  One obvious problem is that it will be difficult to mount
>optical sensors for feedback.  How common is it to find complete DC
>gearhead motors with built in sensors?
>
>Thanks for any suggestions.
>
>Deb Roy

  I am in the process of finishing the drive system for my robot which has
4.25 inch wheels and will weigh about 20 pounds loaded.  I am using a 
differential (Skid turn, I assume you mean) drive with only 4 wheels.
It was brought to my attention that the torque required during turns is
much greater than simple forward motion, and after building a lego technic
platform I proved that this was going to be a problem.
  I changed my gear ratio so it would give me a better advantage and now
that it is basically done, I tested one side (the wheels on the left side
are driven by a vexta (oriental motors) PX244 stepper motor (20 Oz/in (I think))
from a 35 tooth gear to a 128 tooth gear. The other wheel is attached by a
toothed belt to the driven wheel.  The right side gets the same drive hardware.

   I have tested one side with more mass then I expect to need and it was
able to skid turn it around.  My robot is 16in wide and only 10 in wheelbase.
Looks kind of strange, but I don't think I have enough torque to turn it
if I used the 12 inch width of the chassis instead of the 16 inch.

  [ Ascii art deleted,  I can't draw]

   I'll know more when I get my other stepper drive circuit running, but 
from my initial tests it looks like it takes at least 10 times the torque
to spin (forward on the left, backward on the right) as it does to roll
forward.

   Just beware that the skid turn takes a ton more torque than forward motion,
and the heavier robot will have even more problems with this.  I still
may have to redo my drive train and add steering (like a car, instead of a
bobcat.)


--
Brynn Rogers     roger034@gold.tc.umn.edu

