Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!unmvax!pprg.unm.edu!krukar
From: krukar@pprg.unm.edu (Richard Krukar [CHTM])
Newsgroups: comp.robotics
Subject: Re: velocity sensing for robotic joints
Message-ID: <28373@pprg.unm.edu>
Date: 9 Aug 90 22:49:56 GMT
References: <1990Aug7.205751.21206@ecf.utoronto.ca> <19481@well.sf.ca.us>
Reply-To: krukar@pprg.unm.edu (Richard Krukar [CHTM])
Followup-To: comp.robotics
Organization: U. of New Mexico, Albuquerque
Lines: 26


	To sense velocity from an incremental optical encoder,  you only
need a little digital circuitry.  There are two functions to consider,
speed and direction.

	First, direction because it is easy.  Use some edge triggered
flip-flops.  At the positive or negative going edge of one of the quadrature
signals, the other signal's state will indicate direction.

	Next, speed.  A low resolution method would be to feed one of the
signals into a counter.  Use a crystal and another counter to esablish
an "integration time".  Example: a 100 kHz crystal ( clock chip ) feeding
a counter doing a divide by 500 function yields a .005 second integration
time.  So every .005 seconds ( on every tick ) latch the frequency counter
output into a register and then reset the frequency counter.

	Changes to the basic theme can yeild better resolution.  For instance,
create a state machine to read the quadrature outputs.  Note the the output
is a grey code ( 0-1-3-2-0-1-3-2 ) indicative of four states.  If the last
state was three, and the inputs indicate you have entered state one, then
you know the direction of rotation.  Also, on each state transition send
a pulse to the frequency counter.  Now you have a full blown quadrature
detector/speed measurement capability.  Did you specify what this sucker
needs to interface to?

	Richard Krukar ( krukar@chtm.unm.edu )