Searching \ for '[EE]: More motor control.' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/io/motors.htm?key=motor
Search entire site for: 'More motor control.'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: More motor control.'
2000\11\15@204433 by John Mullan

picon face
I've picked two seamingly identical motors with built in gearboxes that I
want to use for my small robot project.

PROBLEM:  When I hook both to the same power source, their rate of rotation
is not identical.  One rotates a little faster than the other. It gains one
revolution in about 10-15 revolutions.

Yes, I will have each motor under seperate 'H' Bridge control.  And I
suppose, for distance measuring, I may use optical flaging.

QUESTION: The circuit will be PIC enabled.  Does anybody have
recommendations on the best way to compensate for the difference in
rotation?  I don't want the robot forever moving in circles.

John Mullan

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2000\11\15@230749 by Jinx

face picon face
> QUESTION: The circuit will be PIC enabled.  Does anybody have
> recommendations on the best way to compensate for the difference
> in rotation?  I don't want the robot forever moving in circles.
>
> John Mullan

How about identical sensors on each shaft and then check for the
arrival of the pulses from each. If one starts to lead, then ease off
its drive until the two are in synch. Finer resolution, more pulses
per rev, like a simple slotted disk and opto, would give less noticeable
adjustment

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2000\11\16@064356 by w. v. ooijen / f. hanneman

picon face
> QUESTION: The circuit will be PIC enabled.  Does anybody have
> recommendations on the best way to compensate for the difference in
> rotation?  I don't want the robot forever moving in circles.

Dissect a mouse and use the two rotary encoders to make your PIC aware of
the exact rotation of the wheels. When no slip occurs this would enable you
for instance to ride one meter forward and then rurn left 90 degrees.

Wouter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email spam_OUTlistservTakeThisOuTspammitvma.mit.edu with SET PICList DIGEST in the body


2000\11\16@092526 by Thomas McGahee

flavicon
face
To compensate for the difference in rotation you first have to have some
means of sensing the rate of rotation. You mentioned that your motors have
built in gearboxes. If you have access to the motor shaft itself, then you
can use a simple "rotation detector". This can be as simple as an addition
to the shaft that has a single hole in it. An LED/phototransistor can then
detect the shaft rotation. I have, on occassion, just slipped a short length
of plastic or aluminum rod over the end of the shaft. A single hole drilled
across the rod allows light to shine through twice per revolution. Because
of the gearing, this still allows considerable accuracy in the control of
the output shaft rotation.

MOTOR-    SHAFT
----------+
         |
         | |------
         --   _   |
         --  (_)  | ROTATING HOLE
         | |------
         |
----------+


Measuring the number of flashes of light detected per second would
give you a good indication of motor speed. (You can also measure
period instead of frequency). To keep the speed of two motors
constant, use one as the reference, and adjust the other so that
it matches the speed of the reference. PWM control works well.
It is a good idea not to run the reference motor at full speed,
just in case the other motor cannot achieve that speed.

If you cannot put a rotation detector on the motor itself,
then you can still put something on the output shaft. In that case
you would probably want to generate as many pulses per revolution as
possible. I sometimes make rotary slot patterns on high contrast
printer's film. I first print the design larger than normal on
my laser printer, and then have the print shop shoot the negative
at reduced size so that I get a very crisp, clean negative.
This is then cut to size and mounted on a thin sheet of plexiglass
that helps hold it rigid. This in turn is mounted somewhere along the
wheel axle and an LED/phototransistor pair is used to detect the
movement of the slots as the wheel rotates.

You can also use a reflective method instead of a transmissive
method. In that case you don't need a film reproduction. Just
print directly on white paper. Mount the paper on a rigid disk
that then attaches to the shaft/axle. When using a reflective
version, it is often useful to use a slit or small peephole
in front of the phototransistor to limit the portion of the
pattern that the phototransistor can see. This cuts down on
false signals.

Fr. Tom McGahee



{Original Message removed}

2000\11\16@093517 by Olin Lathrop

flavicon
face
> QUESTION: The circuit will be PIC enabled.  Does anybody have
> recommendations on the best way to compensate for the difference in
> rotation?  I don't want the robot forever moving in circles.

I don't understand why your feedback loop wouldn't take care of this
automatically.  How are you sensing the rotation speed of the motors?


*****************************************************************
Olin Lathrop, embedded systems consultant in Devens Massachusetts
(978) 772-3129, .....olinKILLspamspam@spam@cognivis.com, http://www.cognivis.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listservspamKILLspammitvma.mit.edu with SET PICList DIGEST in the body


2000\11\16@104353 by Dan Michaels

flavicon
face
Wouter wrote:
>> QUESTION: The circuit will be PIC enabled.  Does anybody have
>> recommendations on the best way to compensate for the difference in
>> rotation?  I don't want the robot forever moving in circles.
>
>Dissect a mouse and use the two rotary encoders to make your PIC aware of
>the exact rotation of the wheels. When no slip occurs this would enable you
>for instance to ride one meter forward and then rurn left 90 degrees.
>


I wonder how they do this in the tanks that Tom was talking
about? -[where are you Tom?] - and in the simple little 2-motor
wart devices sold by Parallax and others. Might be a place to
start looking.

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email .....listservKILLspamspam.....mitvma.mit.edu with SET PICList DIGEST in the body


2000\11\18@132717 by John Mullan

picon face
I haven't done that part yet.  That's part of what I'm asking (although I
may not have been clear about it).  I'm leaning toward some sort of optical
sensing one on each wheel/motor and compensate somehow.  It's important to
note that I have not done this type of interface (motors) yet.  Never
tackled PWM (although I have a good grasp of what it is).

If it helps, I am going to use L293 chip to drive both motors.  No real
current sensing, it's built-in.  So for distance and speed, I like the
optical idea.

John

{Original Message removed}

2000\11\18@132720 by John Mullan

picon face
Say Padre, that sounds good.  Do you have a file you can send of a good
circular pattern that I can work from?

John Mullan
Technical Manager
O.L.G.C.


{Original Message removed}

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...