View Single Post
  #5   Report Post  
Posted to alt.binaries.schematics.electronic
Ian Field Ian Field is offline
external usenet poster
 
Posts: 1,405
Default Very low power dynamo (alternator actually).


"John Larkin" wrote in message
...
On Sun, 27 Nov 2011 14:56:00 -0000, "Ian Field"
wrote:

Tinkering with the old Sturmey archer bicycle dynamo, I was wondering
whether it would charge a 1.2Ah SLA any better if the loading was
modified.

The generator is rated 6V/3W, but off load at a decent rate of knots it
can
produce over a couple of hundred volts.

What I was wondering was whether its possible to get more energy into the
battery by letting the generator output voltage stretch its legs so to
speak
and convert the excessive voltage down with a buck converter.

Can anyone advise on the practicality of this please?

Thanks.


You'd have to experiment with different load resistances to see what
it can deliver.

Alternators like this are often designed with a lot of internal series
inductance. As speed increases, open-circuit terminal voltage
increases, but the series inductance impedance increases too, tending
to make them constant-current sources, ideal for driving light bulbs
at sorta constant brightness at various speeds.

So I'd think that you can get more than 3 watts at higher spin speeds,
with a proper load. A series capacitor, to resonate out the inductance
at some speed, would be interesting.

Given an AC source with internal series inductance, a conjugate load
(resistor in series with a capacitor) maximizes power transfer.

Experiment! Measure!


That would be the ideal route and probably lots of fun, but I don't have
bench space for a jig to hold a hub and provide drive to spin it. And I'd
rather avoid twiddling knobs and reading dials while pedalling a bicycle.

An idea of whether the notion is sound is all I'm really after, if it has a
chance of working I could cobble together a prototype buck converter and see
what happens.

The magnet has a large number of poles so the AC frequency is quite high for
any given RPM, so the inductance doesn't need to be all that high for the
output to approximate to a current source.

The current setup has a pair of largish 6.3V electrolytics from a scrap
motherboard to make the bridge rectifier doubling, getting big enough caps
to handle the 30 - 40V limit imposed by common SMPSU chips might be fun, but
I think the doubling rectifier might be a prerequisite to get a worthwhile
increase from what I'm considering.