Newsgroups: comp.robotics
Path: brunix!cat.cis.Brown.EDU!agate!library.ucla.edu!csulb.edu!csus.edu!netcom.com!jonk
From: jonk@netcom.com (Jonathan Dale Kirwan)
Subject: Re: MicroPower robots
Message-ID: <jonkCrHEoJ.K9J@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
X-Newsreader: TIN [version 1.2 PL1]
References: <2tn8rq$bst@lastactionhero.rs.itd.umich.edu> <2tnnis$dsg@lastactionhero.rs.itd.umich.edu>
Date: Thu, 16 Jun 1994 08:38:43 GMT
Lines: 39

I'd add a few notes.  Definitely take a look at MAXIM parts.  I've seen 
them repeatedly mention an op-amp with 1.2 microamp quiescent current.  
Whatever it drives may increase that some, but...  (It's the 
MAX406/407/409 or the MAX417/418/419.)  Used with a battery, they claim 
to keep running for the shelf-life of your battery (well, almost.)  They 
have a 500 microvolt offset and a 10 femptoamp bias current.  Not too bad.

Also, MAXIM shows a DC-DC converter using their MAX921, a 4093 as an 
oscillator, flip-flop, and buffer, a 10 microhenry coil, a diode, a 
MOSFET, and a few resisters and caps that will draw 2.5 microamps for its 
own use and convert 5 volts in to 15 volts out.  Some adjustments to the 
oscillator and coil may make it effective for your use.  (The solar 
cell's maximum efficiency point, delivering the most joules per unit 
time, may have to be carefully observed and may vary widely with 
different lighting levels.)  The MAXIM DC-DC converter circuit claims 
efficiencies of better than 90% for load currents above 1 milliamp and 
good voltage regulation until 8 milliamps.

One more thing.  Take a close look at the circuit used in flashers for 
marine use.  They have a very simple circuit for boosting a small voltage 
(all the way down to almost a half-volt) to 300v.  I'm not suggesting you 
try 300v, just that a transistor/transmormer oscillator may be a very 
efficient way of pumping the necessary charge onto a capacitor.  It 
certainly can be very cheap.  (Each flash of the xenon tube in these 
marine flashers takes about a quarter joule.  With a claim for about 
30,000 flashes before it stops, when using a single C cell [alkaline, I'm 
sure], that comes mighty close to the number of joules for a C cell -- 
calculated from a book I have.  It must be fairly efficient or the 
sellers must be grossly lying.)

The solar cell's output voltage vs load current curve is just a displaced 
diode V-I curve (moved negatively down the I-axis until the diode curve 
intersects the V-axis at the no-load voltage of the solar cell.)  To 
select the best operating point, look for the maximum V*I.  Roughly, that 
should be at the "knee" of the V-I curve.  This is when the current is 
lightly loaded, often at a voltage about 75% of the no-load voltage.

jonk@netcom.com

