Path: utzoo!attcan!uunet!portal!cup.portal.com!Nagle
From: Nagle@cup.portal.com (John - Nagle)
Newsgroups: sci.electronics
Subject: Re: Can anyone recommend a sensor?
Message-ID: <24648@cup.portal.com>
Date: 2 Dec 89 06:38:22 GMT
References: <472202ca.20b6d@apollo.HP.COM> <69060@psuecl.bitnet>
Distribution: na
Organization: The Portal System (TM)
Lines: 25


     Ultrasonic sonar ranging is difficult for short distances.  The
Polaroid sensor system has a minimum range of about two feet, imposed by
the use of the same transducer for sending and receiving.  You can
improve on this slighty by using active damping of the transducer, I'm
told, but this can't take the minimum distance down to much more than
a few inches.  You can use separate transducers for sending and receiving,
and Cybermation, the mobile robotics firm, has built such devices.  This
gets the minimum range down to an inch or so.  Below that, leakage
between the sending and receiving transducer overpowers reflections.


     There are short-range distance sensors that work on IR reflectance.
The concept is clever; you have one light source and two receivers, set
up as a reflectance system with the two receivers at different distances.
Thus, a nearby object reflects light onto both sensors, but since the
reflected light falls off with the inverse square of the distance to the
object, the levels sensed at the two sensors are different.  Thus, from
the ratio of the values at the two sensors, you can calculate distance.
This works for most diffuse target objects, and was used in a now-defunct
robotics project at Stanford as part of a robot hand to sense the presence
of nearby objects.  Usually, you use a modulated light source and filters
on the sensor outputs to eliminate interference from other light sources.
This works reasonably well.  If you have three receivers, you can do a
consistency check and determine if your data is reliable.


Brought to you by Super Global Mega Corp .com