Hi all

Could any body please shed some light on a circuit to accurately position a DC motor . A PID controller has been suggested by Mr. Nigel here. Any details on encoder construction and application would be of immense help . Thanks always,

__________________
the best bet yet...

 

on some of the old daisy wheel printers the large ones had motors that drove the daisy wheel , very accurately . attached to the motor shaft was a glass type encoder which output a sin wave depending on position...
Just sayin..

 

hi all,

has anyone tried to rip off an encoder from an old mouse (middle click - wheel) . Is it possible . Just asking. :wink:

__________________
the best bet yet...

 

i have not tried it , but it is a good idea...

 

its just an encoder whell ahd encoder wheels are used quite a bit in motor positioning

 

How accurate a position fix does one need?

The encoder from a mouse counting shaft revoultions
of the drive motor , then reduced by a gearbox is
a resonably good method.

Or for the really adventurous the old method employed
by SIP / DIXI is to use a television camera to count the
changes in the interference patterns created by defraction
gratings etched in glass. Accurate to +- to a few microns.

 

An incremental shaft encoder (such as a mouse wheel) only yields position relative to where it was when it was powered up, or when you last zeroed it. Since steering has no home position, how do you know where you are absolutely? Don't you need to know that?
You can get absolute position encoders, but I think they're pretty expensive.

 

Servo's usually use a potentiometer to show the position, as your application requires more than 360 degree rotation (and a pot only does 270 degrees) you could either use gearing from the steering wheel, or (seems a better bet) use a multi-turn potentiometer.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

now that i think of it , the trackball mice have an X-Y setup probably with an encoder on each shaft...just sayin :wink:

 

You could use another smaller motor as a generator. You need to integrate the output of the small motor to calculate position. Accumulation of error would be a problem you'd probably need a fair sized dead band and a opto switch or something to zero out your position occasionaly. This isn't the greatest solution accuracy wise but its cheap and easy to put together. A little bit of gearing to the generator motor to increace its rotation would increace accuracy.

My boss used to design silicon wafer handling machines and they used this method for one of their machines. Apparently it can be quite accurate with high quality components.

 

A stepper motor is the most common solution. As long as you don't force to motor, the position stays in sync with the number of steps. However, you still need a home position detection to know where it is in the absolute sense.