Reclaiming braking-energy via dynamo to charge battery-powered LED lights - circuits

[seanspotatobusiness]

A very clever way to use (two) dynamo(s) friction to slow down AND to charge batteries :
:naughty:
http://cyclurba.fr/velo/78/frein-dynamo-accu-led-velo-eclairage-energie-freinage-recuperer.html

NB: LINK ABOVE IS IN FRENCH - Click here for the translation.

Tally-ho posted this article in a thread that I have regarding charging batteries by dynamo to use for lighting (ultimately took a slightly different direction) and I really like the idea so I've created a new thread for it (I'm not done with the old one yet, and I don't want to make the thread confusing by having too many circuits under discussion at once (as an avid reader of old threads, I tell you it can be very difficult to follow them!)

At first I was revulsed by the use of NiMH batteries, then angered and finally afraid... After barricading the doors, I tried to design a circuit that would use NiCd and super capacitors to handle the surge of charge when braking. I then read that NiMH batteries, like NiCds, handle charging rates of 1C (1xtheir capacity) per hour but since their capacities are much greater than NiCds, they can handle proportionally-higher charging currents. I finally reached the stage of acceptance! E.g. 2000 Ah NiMH batteries (AA) can be charged at 2 A and a dynamo typically produces 500-700 mAh, so such batteries can be used safely with up to three dynamos (I think four would be pushing it).

The problem with the author's design, as I see it, is that the LEDs are connected directly to the dynamos when the circuit is closed and the operator brakes. Fast charging NiMH cells requires a potential difference of 1.4-1.6 V, so when the two dynamos are trying to dump their amp, they'll need to generate a potential difference of 5.6-6.4 V in order to do so through the batteries? My preferred low-power LEDs have a max allowable forward voltage of 3.8 V (which is pretty hearty, if I'm not mistaken) so in order to drop 5.6 V to 3.8 V, with a current of 20 mA, the resistor needs to be 90 ohms. If the voltage across the batteries at 1 amp is even higher, then the resistors should be higher still. In short, I think the original design blows [LEDs]!

My modified circuit uses four dynamos [in parallel] and three batteries in series. I will probably use 10 000 mAh D cells when I get around to building this so I could eventually upgrade to sixteen dynamos

When the switch is closed, and the brakes are off, the 15 ohm resistors drop the 3.6 V to 3.3 V, allowing a jovial 20 mA through each jolly little LED. When the switch is closed and the brakes are on, the voltage is liable to exceed 4 V. For such eventualities, I have designed the Sean Transmogrifier Bridge, constructed from five 0.7 V and one 0.5 V diodes. At 4 V, the LEDs are still operating below their maximum 3.8 V (the drop across the 15 ohm resistors should actually be slightly more than 0.3 V) so they will not be damaged. At 4V, the STB opens, redirecting the current back to sender and preventing damage to the LEDs.



Any questions?

I have one! When the voltage pulses over 4V and the circuit is shorted, will the LEDs still light? They should present a path of lower resistance than the series diodes, right, allowing excess current back to the dynamo?

Oh, oh! At a fixed voltage, say 1.5V, will a 10 000 mAh D allow more current into it than a 2000 mAh AA? IF it did (even if it doesn't) would that mean that it must have a lower internal resistance?

 

[seanspotatobusiness]

It's been said that the system would not sufficiently charge the batteries, nor contribute much to braking, but I disagree and think it depends on the design and the riding-style. Obviously, you can't use power-LEDs, but, depending on how long you cycle at night vs during the day, I think four dynamos can provide useful energy. It depends on a person's riding style, but I ride mostly in urban traffic, rather than long stretches of country road and prefer to brake well in advance of anticipated stops (lights gonna change to red etc). By pulling lightly, the bracket allows contact of the dynamos before the brakes. Four cheap dynamos will surely provide noticeable drag, if applied at the same time, which is fine for anticipated stops, most of which, for my same ole, same ole road journeys are.

If I generate 2 A, pulling on the brakes, and use 200 mA on my lights, that's 10 times as long lighting as I spend braking and given that I've ridden only once after dark since I got my bike a couple of months ago, I think it will work. Admittedly, it's summer now and in the winter, the ratio certainly changes, but then the batteries can always be removed and charged, but with much lower frequency that without this system.