Newsgroups: comp.robotics
Path: brunix!news.Brown.EDU!noc.near.net!das-news.harvard.edu!cantaloupe.srv.cs.cmu.edu!crabapple.srv.cs.cmu.edu!e4.ius.cs.cmu.edu!hbb
From: hbb@e4.ius.cs.cmu.edu (Ben Brown)
Subject: Re: Muscle wires for robots ?
Message-ID: <CA2I6M.50x.1@cs.cmu.edu>
Keywords: SMA, muscles, wires, 
Sender: news@cs.cmu.edu (Usenet News System)
Nntp-Posting-Host: e4.ius.cs.cmu.edu
Organization: CMU Vision and Autonomous Systems
References: <26401@mindlink.bc.ca> <CA0Bvt.3q6@world.std.com>
Date: Mon, 12 Jul 1993 19:59:03 GMT
Lines: 23

In article <CA0Bvt.3q6@world.std.com>, rgacote@world.std.com (Raymond Cote) writes:
> Jeff_Holinski@mindlink.bc.ca (Jeff Holinski) writes:
> 
> >> William writes:
> >>
> >> Ok, how do you manage "anywhere", when the muscle wires themselves are
> >> basically digital (long/short) ?
> >>
> >Couldn't you use a 555 timer to switch the power off and on? The more pulses
> >per second you use the more the wire contracts.
> >Anybody tried this?
> Roger Gilbertson out at Mondotronics is using a PIC controller to do 
> pulse-width modulation control of his six-legged robot. Light, cheap, and 
> easily programmable.
> ray cote
> rgacote@world.std.com

Aside from slow response, another problem with SMA (shape memory alloy)
actuators is inefficiency. Typically they produce mechanical work less
than 10% of the input energy. I have heard figures of a few percent
quoted. I don't know what the theoretical limits on effiency are.

	Ben
