Can a Simple Relay Prevent Damage from High Voltage in a Hub Dynamo?

[colintonks]

Hi,

I am wondering if you can help:

I have a hub dynamo that I would like to use to charge my mobile for a cycle ride across europe.

The DC hub dynamo produces anything up to 40volts at full pelt!

So I want to use a car charger to charge my mobile phone (and one for a camera), which I assume takes a maximum of about 14volts.

Problem is that any voltage >14volts will surly fry the car charger.

So is is possible to have a simple circuit, say some form of relay that will 'open' the circuit at >14volts thus disconnecting the car charger and preventing damage.

I have had a good look around the web and yes I have found some stepdown transformers, however many of them don't handle high voltage and are also expensive.

Is there a simple solution, i.e. some sort of relay or am I clutching at straws!

Thanks
Colin

 

[sophiecentaur]

What you need, ideally, is a voltage regulator.
There are chips that will do the job for you and the data sheets often contain details for including them in a circuit.
I don't know how much current your dynamo will produce - are you sure it's not AC (? they always used to be because lamps aren't fussy and there are no diode losses) so you may well need to bridge rectify its output, in any case. It is probably fairly limited so you may get away with a Zener diode to protect your phone. You'd need to measure the actual current you have available and buy a beefy enough Z diode which will not fry whilst dumping surplus current. A small series resistor could also help. You'd have to experiment with a meter and hang the bike up on a hook and turn the pedals to test (carefully) what happens.

 

[Adjuster]

N.B. The following description assumes an AC "dynamo" which almost all bike generators are, so please ignore this if you really have a DC model.
These things depend on having a relatively big internal (leakage) inductance to achieve a crude sort of regulation against speed variation when connected to an appropriate lamp bulb load. The idea is that as the speed and so voltage increase, so does the frequency, and the increasing internal reactance of the "dynamo" restrains the increase in lamp current. A corollary of this is that the load regulation is poor. Hence the whopping great open-circuit voltage.

There is an upside though: it very likely WOULD be practical to use a shunt zener after the rectifier, although it would need to be rated (and heat-sinked) for a few watts.

The big effective series inductance also means that this type of generator is not great for feeding a rectifier into a smoothing capacitor - you might not get as much current as expected. Note also that your dynamo may well be designed for a nominal 6V, so that the current yield into a rectifier may not be enough at modest speeds. In this case, you could try a voltage doubler rectifier circuit feeding the Zener, instead of an ordinary bridge. Don't omit the Zener though, or you'll get 80V!

 

[sophiecentaur]

If it really is AC you could use a small transformer to increase your volts a tad, so that you would get what you need after the rectifier. A pretty easy spec for a transformer, I reckon.

Yes - I remember an A level question involving the L of a cycle dynamo and the regulating effect it would have as the frequency went up.

 

[Adjuster]

The minimum frequency may be below the 50 or 60Hz most small power transformers are designed for, so they might not have enough inductance to work well at lower speeds. I seem to remember the bulbs on my student bike flickering visibly at low speed - but then again I never was much of an athlete.

You could use a transformer having windings meant for a bit more voltage, say >20V, or perhaps use an audio transformer? Similarly, if you did use a diode-cap voltage doubler, the capacitors would need to be big, a few thousand microfarads should do.

 

[Adjuster]

Hello colintonks,

You may wonder why I suggested upping the voltage, it all depends on how much current you can get and at what speed.

Incidentally, you really need to have an estimate of how much current the charger takes from a 12V supply, while charging a run-down battery. Can you measure that, or is there a handbook figure?

In any case, your starting point is to resolve the DC/AC question once and for all. Then, if it is AC attach a suitable rectifier and smoothing before proceeding as sophiecentaur suggested in his original post. (Perhaps you already have those fitted?)

It might not be a bad idea if you could confirm whether these ideas seem to make sense?

 

[colintonks]

Thanks everyone for your comments!

I am amazed just how much information you have provided in such a short amount of time.

Sadly, I am well out of my depth (as Aduster alluded to) and don't really understand how all you comments summarize into a simple circuit. You were all correct about the hub being AC, in fact it is this one:

http://bike.shimano.com.sg/media/te...V0G-DH-3N30_3N20-EN_v1_m56577569830600092.pdf

If it is not too much to ask, could someone please summarise a simple circuit that a complete novice could build!? Id also be happy to buy something off the shelf if possible.
Thank you!

i.e. will this do it:

http://i143.photobucket.com/albums/r154/n4zou/lights/DynamoUSB-2.jpg?t=1252975944

 

[Adjuster]

Thanks everyone for your comments!

I am amazed just how much information you have provided in such a short amount of time.

Sadly, I am well out of my depth (as Aduster alluded to) and don't really understand how all you comments summarize into a simple circuit. You were all correct about the hub being AC, in fact it is this one:

http://bike.shimano.com.sg/media/te...V0G-DH-3N30_3N20-EN_v1_m56577569830600092.pdf

If it is not too much to ask, could someone please summarise a simple circuit that a complete novice could build!? Id also be happy to buy something off the shelf if possible.
Thank you!

i.e. will this do it:

http://i143.photobucket.com/albums/r154/n4zou/lights/DynamoUSB-2.jpg?t=1252975944

Not quite, I'm afraid - it does show a rectifier bridge which is where you would have to start, but basically what you have there is a rather crude way of powering a USB type outlet to about 5V. The voltage stabilisation element is an NIMH battery, so the output voltage would not be that consistent. There would also be a considerable chance of overcharging the battery on a long run.

Actually, it occurs to me that if you could get chargers designed for 5V USB input, these might be a much better match to the dynamo output. Such chargers may also be more efficient - I think some in-car chargers are quite poor in that respect.

I believe that something along the lines of rectifier - smoothing - Zener limiter - series regulator (or switcher) could be made to work. It would however be quite a little development job in itself. I don't know of a tried-and tested circuit for doing this, and frankly I would rather not try to dream something up for you which might not work out in practice, and could damage expensive equipment.

Don't give up just yet though, somebody else may well have some better ideas, or at least more time to think about them. Sadly, I've got to go to work!

 

[sophiecentaur]

Have you thought of trying / joining a cycling / touring forum?
I bet you're not the only person who would want to charge a mobile on the road. Someone may have solved your prob already.

 

[sophiecentaur]

Hey - look at this.
http://www.cyclechat.co.uk/forums/showthread.php?t=8491

 

[Bob S]

You didn't mention what your current requirement or limit is. I believe the best choice is to use a bridge rectifier to create a dc voltage. There should be a capacitor on the output to provide enough energy storage at low wheel RPM. Then use a dc SEPIC (single ended primary inductor converter)(boost/buck) regulator, with current limit.

This might help find a suitable circuit

http://parametric.linear.com/html/sepic_regulators

Bob S

 

[sophiecentaur]

Bob S
I don't think you would need a smoothing capacitor. It may produce a better / smoother looking Voltage waveform but would it necessarily improve the charging of the battery (which will more or less maintain its own voltage, in any case)? At low revs, I doubt there would be more than the occasional small dollop of charge flowing at each peak of the cycle (the emf will drop at low revs). Indeed, it could even be better not to use a capacitor which might even reduce these peaks.

 

[Adjuster]

Bob S
I don't think you would need a smoothing capacitor. It may produce a better / smoother looking Voltage waveform but would it necessarily improve the charging of the battery (which will more or less maintain its own voltage, in any case)? At low revs, I doubt there would be more than the occasional small dollop of charge flowing at each peak of the cycle (the emf will drop at low revs). Indeed, it could even be better not to use a capacitor which might even reduce these peaks.

I'm not sure about that. I suppose whether or not you would need a reservoir capacitor would depend partly on how the battery is to be charged. A proprietary charger designed for 12V or 5V DC input might not take very kindly to unsmoothed input.

In fact, a reservoir BATTERY might be a better way to deal with this, together with an output switch with a hysteresis arrangement to allow the reservoir battery to charge before switching on the output, switching off if the reservoir battery brecomes depleted. Otherwise the phone charger could power up and down repeatedly and rapidly as the bike speed fluctuates, which could result in malfunctioning.

It might be possible to omit smoothing for charging some types of batteries directly, but to make a safe charger for small phone batteries would be more complex. Remember that the enquirer is not expert, and the batteries may well be Lithium types, which need careful handling.

 

[sophiecentaur]

I think the experience which is quoted in post 10 may be useful. There's no point reinventing the wheel. If someone has actually made a system which works long-term then that seems to be the way to go.

 

[Adjuster]

Agree with that, despite being unable to understand the linked page in German. It would be interesting to know what the arrangement is, but of course the owner may not wish to have those details posted.

 

[sophiecentaur]

I should never dream of suggesting anyone should stop waffling! Life would not be fun if we couldn't do that!

 

[dlgoff]

If you look at Bobs link, you will see that the regulators he was suggesting require a minimum input voltage, hence the need for a capacitor.

 

[colintonks]

I think the experience which is quoted in post 10 may be useful. There's no point reinventing the wheel. If someone has actually made a system which works long-term then that seems to be the way to go.

Thanks 'sophie'!

I think you are absolutely right, especially given my limit to no experience in this area! I'm going to follow up on the lead that you have presented here and want to say thanks to everyone!

I'll keep you informed!