> Yes. I might stop eventually, as long as it can move up to 1 hours or 2
> hours. It is consider quite successful.

AND

Roman said:
> Don't let him discourage you James. :o)
> This sounds like a great hobby project and a fun
> way to learn a LOT about electronics and ohms law
> etc. And you may have an easier time suceeding if
> you allow a little energy input from heat or light
> etc, but never believe someone who tells you it can't
> be done... That just means HE couldn't do it.
> :o)

Don't take Roman's comments too much at face value. DO please notice the
smiley face at the end of his comment. He wasn't making fun of you and you
certainly WILL learn a lot about electronics and you certainly will be
helped by extra input from eg heat or light BUT you will have an
extroadinarily hard task trying to make something continue working for an
hour or two from simple brief human input.

As several people have said - having an alternator etc drive a motor to
drive the alternator and so on is a form of perpetual motion machine and
will NOT produce any extra energy. Even Roman could not make it do so :-).
Worse, trying to do it this eway introduces losses which are hard to
minimise compared to alternatives.

Consider a streamlined flywheel of several kg mass running on precision
bearings and with an aerodynamic design. Wind this up to high revs by hand
in some manner and then leave it to run. As you can imagine, very few such
designs would turn for more than 5 to 10 minutes even if initially run up
"very fast". Try it with a push bike wheel - not very streamlines but quite
a good example. I haven't done this for years but I think you can get
several minutes.

IF you are happy for the person to initially turn eg a crank quite hard and
fgast for some seconds then you may well be able to get a well designed
oscillatory circuit to work for the sort of period you have in mind.As an
example, some of the rocking dolphin/man/golfer/sphere/whatever devices you
see as desk toys will run for extended periods on a single cell. Lets say
that one such can run on a single alkaline AA for a year (don't know how
realistic this is). A year = 8765 hours. An AA Alkaline has about 3000
milliamp hours capacity so the average current = 3000/8765 =~ 1/3 millaimps.
This is quite high! Say average battery voltage = 1.2 volts so average power
= 1.2V x 1/3 mA = 0.4 mW. That's not much power. If you want this to run for
say an hour = 3600 seconds then you need 3600 x 0.4/1000 = 0.9 watt seconds.
Say 1 watt second. The user only needs to input about 1 watt second !!!! A
watt second is expended by a force of 1 kg travelling over a distance of 0.1
meter in a  second. Turn a crank with a 15 cm throw one turn in one second
with a force of 1 kg provided by an alternator and you have input a watt
second. Turn it twice in 2 seconds to allow for inefficiencies.

Storage is more formiddable. Mechanical storage may be easier. There are
clockwork wind-up radios availaable which would meet this requirement
extremely well.

Lets look at a capacitor solution though. We want 2 watt secons storage =
2000 mW = 2000 ma V storage. If we used a switching regulator to transform a
higher voltage down to the oscillating device voltage (here = 1.2v say) we
can get a much better result. Lets assume a 10 volt storage voltage to start
and a linear current discharge and a switch mode power supply (SMPS) with
100% efficiency for now.
Energy = 0.5 C V^2 or C = 2E/V^2
At 10v, 2 watt seconds. C = 2 x 2 / 100 = 4/1000 Farad = 40,000 microFarad
!!!
Rather large. (I have an 85,000 uF at 16V here and it is a full handful.

By going to say 100 v storage we need 2 x 2 / 10,000 = 400 uF.
Much more manageable BUT leakage is liable to be a problem.

That's enough to give you some ideas.
Spring storage is MUCH easier to get good energy density.
Even pumping water and having a nano-hydro scheme might be practical :-)
(A watt second = 1 kg at 0.1 meter or 100g at 1 meter or ....)

Having a REALLY efficient oscillating mechanism may be practical.
The 1/3 mA I mentioned seems excessive for a really well balanced low drag
device.
0.1 mW will move 1 milligram  100 cm. If you can get your imbalance and drag
forces down to the milligram level then you can deal in submilliwatt power
levels.

As Roman noted a very very small amount of solar cell input here would be of
great assistance.
To get your 0.4 mW power input you need a solar cell about 4mm square !!! in
bright sunlight and proportionately more as the light level falls.
Get a cheap solar cell from an eelctronics shop or take one from a dead (or
alive) solar calculator. Measure open circuit voltage and short circuit
current uin roomlight and sunlight and then load it with various resistors.
Get an oswciallting desk toy and see if one can be solar powered by this
panel. (Use  a small elctrolytic capacitor across the panel output as these
devices draw current in bursts as the swinging magnet passes.

A VERY strong but relatively light rare earth magnet on the bottom of your
oscillating device could produce enough energy to a coil beneath it IF you
areprepared to have the user puit enough force into it to start. An easier
approach is to have one oscillating device with a heavy weight and a rare
earth magnet plus a second oscillating device optimised for low current long
life oscillation. The user swings the heavy device to make power and thr
light one then runs for along time. Storage still a problem as above.

Can we ask what this device "does"?


Have fun.


        Russell McMahon

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads