Newsgroups: comp.robotics
Path: brunix!sgiblab!spool.mu.edu!agate!doc.ic.ac.uk!uknet!demon!rednet.demon.co.uk!marc
From: Marc Nicholas <marc@rednet.demon.co.uk>
Subject: Re: Fiber optics to determine bending?
Message-ID: <CIyp8K.MFC@demon.co.uk>
X-Xxmessage-Id: <A94B664672060920@rednet.demon.co.uk>
X-Xxdate: Sat, 1 Jan 94 17:41:26 GMT
Sender: news@demon.co.uk (Usenet Administration)
Nntp-Posting-Host: rednet.demon.co.uk
Organization: RedNet Ltd. - The Mac Networking People
X-Useragent: Nuntius v1.1.1d12
References: <2fcu39$nu8@atlas.cs.upei.ca>
Date: Sat, 1 Jan 1994 17:43:31 GMT
Lines: 33

In article <nagleCIMJtq.5CG@netcom.com> John Nagle, nagle@netcom.com
writes:
>      Fiber optic transmission changes with bending.  Tight bends,
>though, damage the fiber.
>
>      VPL's initial version was even simpler; it used plastic tubes with
>photodiodes at one end and LEDs at the other.
>
>      The Mattel PowerGlove used a conductive plastic placed so that the
>strips on each finger elongated as the finger bent, changing the
>resistance.
>
>      Strain gauges have also been tried.

Using fiber optic light guides (available from Radio Shack) you can build
these sensors.....I actually cut through the fiber (!) leaving one side
of the insulator intact....then cover this will opaque material
(insulating tape works well)...this will give you about 50 degress of
sensing....however, it's not paricular elegant or accurate, but
demonstrates the point. By pointing one end at a light source and looking
at the other end you can actually *see* the light level changing.

The PowerGlove sensors were not all the same...I seem to remember Mattel
having to change the sensors coz the original supplier stop making them
or something...the PowerGloves I've got use standard resistive bend
sensors that do *NOT* elongate...


L8R,



Marc.
