To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.robotics.handyboardOpen lugnet.robotics.handyboard in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Robotics / Handy Board / 528
527  |  529
Subject: 
Re:
Newsgroups: 
lugnet.robotics.handyboard
Date: 
Wed, 26 Jun 1996 22:40:30 GMT
Original-From: 
Chuck McManis <CMCMANIS@NETCOM.nospamCOM>
Reply-To: 
cmcmanis@netcom.com^AntiSpam^
Viewed: 
1597 times
  
Lyle Hazelwood wrote:
My earlier posting has left me thinking.. (Suprise!)
A MUCH more efficient way of doing PWM would be to use another 'HC11.

Actually an 'HC11 isn't a very efficient PWM controller. However a
Microchip PIC is. (or a Motorola 68HC05, or a Siemans 87C752, etc)
These single chip microcontrollers are both cost effective and
easy to use.

In the July (or August) '95 issue of Circuit Cellar Ink I published
an article on using the cheapest PIC to control an H-Bridge (and
motor) The advantage was that the PIC could easily give me 100
different power ranges whereas the software solution on the HC11
gave me 8 to 16. I'll see about putting the source code up on my
home page.

Now in my article I used a one-bit interface based on the Servo
interface that most servo controllers use. This was useful because
I already had a servo library for the miniboard but is less useful
on the Handyboard given the number of lines on the base board.

However, also in that issue (and the next) one of the columnists
talked about using one of the beefier PIC chips to control two
motors and to use the I2C bus (which is similar to the SPI bus).

Since the typical implementation of a PIC computer is four parts
(PIC, ceramic resonator, two resistors.) These things are pretty
easy to use. Also there is a company in Colorado that sells a
pretty cool "PIC Proto" board which has a PC layout for the PIC
and some breadboard area for the you to use.

If the redesigned handyboard expansion board continues to support
a PIC (I vote for it ;-)) Then you can do this with the expansion
board as well.


Some Notes About Motor Control
==============================

Of course this is all well and good, except that we have a slight
problem of motor control semantics.

One of the interesting things about motor control is that is
can be simplified usefully to "motor on/detect/motor off" but
when you want to start doing precise things your looking at
"motor on/motor to speed/distance covered/motor off" and that
required a bit more sensing of the environment.

I originally designed the PIC motor controller because I needed
finer control over the motors in my "Wild Cougar" robot. These
motors are very powerful and go very quickly. Thus to control
them you need a fine control over the power applied.

If you build this motor controller you will find that in general
speed != power applied, it isn't even the simple function
speed = x * power. No, its much more complicated. Further when
you add a feedback network to control speed (usually in the form
of encoders) you will find that correctly built (ie critically
damped) feedback systems will take finite time to ramp from
zero to the desired setpoint. Calculating distance travelled
then becomes somewhat more difficult than simple speed * time;
its more like the integration of the speed function over time.

One way to finesse this problem is to not only take the first
derivative of your wheel encoders (as described in the mobile
robots book) but also to accumulate the total pulses to understand
how "far" your robot has travelled. If your robot operates with
two wheels driving (differential steering) then you will need
to include differential effects (the outer wheel travels farther
in a turn.)

Anyway, you might find yourself back to using an HC11 for the
task :-). National Semi makes a chip that does all of this tracking
but they are $90 each.

I've not yet found a really good compromise that has the level of
return for the amount of effort that the "turn 'em on, turn 'em off"
design has. Another alternative to this issue is stepper motors.


--
Chuck McManis                    http://www.professionals.com/~cmcmanis
Director System Software, FreeGate Corp.            cmcmanis@netcom.com
All opinions in the non-included text above are those of the author and
not of his employer or avocado plant.



1 Message in This Thread:

Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR