Hi,

I would appreciate if you could give me some advice on my new project.
It's simple.

Bicycle dynamo (6V,3W) -> Diode Bridge (with 4700μF 35V capacitor) -> Buck Boost Converter -> Mini USB Lithium Battery Charger -> Lithium Ion Battery (3,7V 900mAh)

However I have some questions:

1.) Is 4700uF smoothing capacitor a bit of overkill?

2.) Dynamo output is fixed at ~500mA, but what concerns me is when I'm riding downhill voltage can rise above 20V. Capacitor and buck-boost converter are rated at 35V max... Do I need some kind of protection circuit for short intervals of higher voltage? If so, what would you suggest?

3.) I will connect lamp to the li-ion battery with really thin cables (for electronic projects purposes, don't know the gauge). Can they withstand 1A from battery?

Thank you!

 

The dynamo will give out dc voltage at 20V max as you said downhill, so you wont need a bridge rectifier, just a single diode in series with the output of the dynamo, like an IN4001 to prevent feedback into the dynamo.

The buck converter pcb will take 35V input max, so 20V isn't a problem, any electrolytic capacitor 35V from 1000uF to 2200uF should be ok.

However if you measure the voltage with a DVM and see if its DC or AC to be sure, then if its AC you will need the bridge rectifier, but the supply when rectified with a smoothing capacitor will go upto 28Volts.

 

The dynamo will give out dc voltage at 20V max as you said downhill, so you wont need a bridge rectifier, just a single diode in series with the output of the dynamo, like an IN4001 to prevent feedback into the dynamo.
The buck converter pcb will take 35V input max, so 20V isn't a problem, any electrolytic capacitor 35V from 1000uF to 2200uF should be ok.

The problem is I don't know how much volts could dynamo put out when going downhill. I assume 20V or more. And then because of the capacitor you got to multiply that by 1,4. But, let's say, if it is >35V what is the best solution then. Zener diode? comparator with relay? etc..

Dynamo puts out AC voltage btw. I want full wave rectification.

 

The problem is I don't know how much volts could dynamo put out when going downhill. I assume 20V or more.

Without much load, and within a range of speed, the voltage is going to be linearly proportional to speed. You may be able to do a few experiments to plot voltage versus bicycle speed.

Why do you think dynamo current is limited to 500mA? I'd expect it to rise with speed also.

If the voltage is not excessive until you exceed, say, 40 mph, I'd say it's no worry. But maybe you'll need a circuit to protect against over-voltage. It would be easy to dump excess power (maybe with a zener to the base of a transistor that controls the dump current), but this would have a braking effect on the dynamo and bike. Maybe that would be good? A cutoff that kicks out the converter would avoid that braking, but would turn off charging right as you have peak charging power available. Maybe it would be rare enough that you just wouldn't care about losing a few seconds of charging.

 

Without much load, and within a range of speed, the voltage is going to be linearly proportional to speed. You may be able to do a few experiments to plot voltage versus bicycle speed.

Why do you think dynamo current is limited to 500mA? I'd expect it to rise with speed also.

If the voltage is not excessive until you exceed, say, 40 mph, I'd say it's no worry. But maybe you'll need a circuit to protect against over-voltage. It would be easy to dump excess power (maybe with a zener to the base of a transistor that controls the dump current), but this would have a braking effect on the dynamo and bike. Maybe that would be good? A cutoff that kicks out the converter would avoid that braking, but would turn off charging right as you have peak charging power available. Maybe it would be rare enough that you just wouldn't care about losing a few seconds of charging.

I have ordered bicycle computer but it hasn't arrived yet. When it arrives I will make U/speed characteristic. A lot articles on the internet say that bottle dynamo is designed to limit the current to around 500mA. I tested that with multimeter and it is true. But I didn't go over 15V so I might do that again to see how current acts at "higher" voltages.

I don't mind losing few seconds of charging but I want to avoid destruction of components at all costs.

Found this high voltage cutoff on the internet. Is this schematic good?

 

It would work, although the "output" would not be capable of doing anything on its own. You'd need to drive a MOSFET switch to control the current through the rest of your circuit. If you do go to build this, use a comparator such as the LM393 (dual) or LM339 (quad). The 741 op amp is literally the oldest you can find and not ideal for anything.

I'm thinking of something like this (along with the input arrangement of your circuit). The feedback resistor is optional but is used to give some hysteresis so that the output doesn't chatter at the transition point.

 

Thank you!