Cheap dynamo USB power

[edocaster]

Hi,

Has anyone seen this? http://www.ebay.co.uk/itm/Universal-New ... 1036449456

It looks a bit like a 'Spinpower S1' - http://www.bike2power.com/smartphone-bi ... r-kit.html - just ridiculously cheap.

My question is, would either device run properly from a hub dynamo? Or would they self-destruct? I have little understanding of how the output of hub dynamos work, but keep on reading about other, more expensive devices failing on fast descents - i.e. the voltage of the hub can't be kept down? Is this a problem which only affects hub dynamos and not bottle dynamos?

[jb]

Hi
The divice should regulate the supply from any 6v bicycle dynamo, hub or otherwise. Neither bottles or hubs regulate their own power not the common ones anyway.

[Brucey]

there is quite a lot of data on the characteristics of various generators out there.

Very many (both hub and bottle) are designed in such a way as the no-load (or low-load) voltage simply increases almost proportionally with speed. Since they will typically make a reasonable effort at 6V on load at just 8mph and manage 6V off load at a somewhat lower speed you can see that 30-50V is possible at speed with some generators when there is no load present.

Since a regulator typically reduces the load in order to regulate the voltage, the components in the system need to be able to withstand voltages of at least 50V in order to be proof against all generators under all conditions. Not all systems are so rated, hence the failures.

If you buy an e-bay cheapie you might get lucky, you might not. Even good quality systems have occasional QA problems, so I'd suggest that you buy good quality kit with a good warranty here, especially if failure isn't an option (e.g. because it will leave you without lights or perhaps even damage any gizmo you have plugged into the generator system).

cheers

[[XAP]Bob]

I run a biologic reecharge (also sold under the dahon brand) as they were on sale for £25 a while ago.

Works well, ut there were reports of the regulator failing in no load scenarios - I leave my rear light always on so I can't hit that failure mode.

[edocaster]

I run a biologic reecharge (also sold under the dahon brand) as they were on sale for £25 a while ago.

Works well, ut there were reports of the regulator failing in no load scenarios - I leave my rear light always on so I can't hit that failure mode.

Interesting - £25 with the power pack? I see that you can get the regulator on its own for £22: http://www.ukbikestore.co.uk/product/23 ... cable.html ...but would it be compatible with any USB power pack, or only the Biologic one? Could it even power a device directly (sounds risky, but surely there wouldn't be a point to selling the device on its own?)

I'm also slightly worried after looking at reply 12 here https://yacf.co.uk/forum/index.php?topic=40524.0 - the output seems to say '6v' - surely too high for USB spec? The manufacturer's page is even more vague ('5-6v', and gives the current as a measly 250ma): http://www.miniwiz.com/miniwiz/products ... /reecharge

So, in short, it seems all these devices have to deal with a wildly fluctuating dynamo voltage which can rise well beyond the nominal 6v AC. I read about some homebrew solutions, but they are all beyond me. I'm really surprised no-one sells a basic in-line device to protect the likes of the Reecharge, as being unable to just leave the device on the bike without a load seems mystifying. The price difference between a basic dynamo light (e.g. Axa Pico 30 - around £18) and one with a built in USB output (Axa Nano 50 Plus - around £58) is still a bit steep though, especially as the nicer lights will always be a theft risk.

[[XAP]Bob]

Yes, with the cache battery...

"load" is provided by a rear dynamo light, it doesn't need to be alot.

The rectifier supplies 6V to the (clearly somewhat basic) electronics to charge the battery - which then provides 5V out...
The rectifier doesn't have a USB output, but a standard round DC plug

[pete75]

Nokia do something similar. http://www.amazon.co.uk/Nokia-DC-14-Cha ... B00475BGHA Personally I'd trust a Nokia product as more likely to deliver reliability and correct voltage.

[jb]

The B&M E Werk works perfectly, gives a choice of voltages and is fairly compact. I've never disconected it even when not in use. I charged my phone, AA batteries & a power pack on a recent tour. It makes the lights slightly dimmer if charging at night.

[Brucey]

"load" is provided by a rear dynamo light, it doesn't need to be alot.

Further to that you can get an idea of how the voltage varies with load by looking at the data here;

http://www.enhydralutris.de/Fahrrad/Beleuchtung/node75.html#2280

which is for a shimano NX-30 generator. [there is lots more data on this site but it is for generators and lamps that are all some years old now]. Testing at a 12 ohm load simulates how much power is produced into a traditional '3W' tungsten bulb load. Testing at 25 ohms simulates (roughly) what will happen with just one light from two connected so it is a bit like what will happen with just a rear light connected. Open circuit (0.8M-ohm) load shows how many volts will be generated with no load connected.

So into a 12 ohm, 25 ohm, and open circuit at 32 kph (~21mph) gives

6.6V, 13.8V, and 28.2V respectively

However down a steep hill at ~65kph (~43mph) will produce ~51V open circuit which is enough to trash poorly regulated chargers etc. In theory if you go fast enough you can get to over 100V with this generator.

Other generators vary in detail but they all show the same basic characteristic of increasing voltage with reduced load and increased speed.

cheers

[[XAP]Bob]

I have done just shy of 50mph and my rectifier/regulator are fine

[edocaster]

Thanks for all the replies. Always an education.

The link to the Reecharge without power pack says 'Directly charge devices via micro-USB' - so this version is different? Hopefully the voltage actually is USB spec.

As for the Nokia device, that seems to have been made for a bottle dynamo, so maybe wasn't designed for the higher voltages of hub dynamos at speed (if they do produce higher voltages)?

A few searches turn up some other devices like the Kemo M172N (the N version appears to be new, and rated to 70V) and the LightCharge, both around the £40-50 mark. Whereas the high voltage characteristics of the likes of the Axa Nano 50 Plus are also unknown (in fact, I believe the switch on that light only allows a choice between charging or light, so presumably a 'no load' situation is quite a likely scenario).

I guess the USB side is just a toy option right now that I wouldn't want permanently attached to the bike, so I might just go for basic lights at first.

[pete75]

I guess the USB side is just a toy option right now that I wouldn't want permanently attached to the bike, so I might just go for basic lights at first.

Or lights and USB in one

http://www.rosebikes.co.uk/article/axa- ... aid:642351

http://www.rosebikes.co.uk/article/b--m ... aid:620049

[edocaster]

I guess the USB side is just a toy option right now that I wouldn't want permanently attached to the bike, so I might just go for basic lights at first.

Or lights and USB in one

http://www.rosebikes.co.uk/article/axa- ... aid:642351

http://www.rosebikes.co.uk/article/b--m ... aid:620049

Indeed. Best price I've found for the first light (may help other forumers):

http://www.amazon.co.uk/Luxx-steady-939 ... namo+light

(The picture doesn't show the USB version, but the text does refer to it.)

Was very, very tempted. But on reflection, for a light which may be left out in the city, I think I'm better off with a 30 lux 'dumb' light, and then having a go at one of the homebrew USB devices. Clearest guide I've found so far is: http://www.14degrees.org/en/?p=5203 ...seems to adhere to the most basic design I've seen around. Technically, won't be much of an improvement on a Biologic Reecharge, but it's all about learning!

One question for the sparkies here though: Some of the parts (e.g. the 2200uF capacitor and the bridge rectifier) I seem to be able to find components with higher voltage ratings for not much more in cost (possibly a little larger in size). Is it worthwhile using those higher rated components? Or does it screw up something else in the circuit?

[edocaster]

OK, quick update. I built the charger - pics below show the board, and the final unit in a vitamin tube (which has far more room than is necessary).

Total cost in parts probably less than £15, not including soldering iron and time!

The part permanently attached to the bike is a pair of wires (recycled from an old phone charger) parallel to the headlamp wires going to the hub (hint: twisting two pairs of wires and shoving them into the Shimano 'lego brick' connector is really difficult - must remember to pick thinner wire next time...), terminating in a Tamiya connector zip-tied to the stem. The charger can be detached when not needed, or when it risks being overloaded.

Only had a brief test today, using an old smartphone (HTC Hero) as test dummy. Started out at 7/10 bars of charge. After 2km... 7/10 bars However, that's no surprise, as the site I followed estimated 1-2% per km, plus as the phone was on, the screen would switch on every time it started or stopped charging (which, incidentally, began at around 8km/h). I'll do a longer test, perhaps with the phone off, soon.

So far, pretty satisfied. If it works as planned, I'll seal the unit and add some more heatshrink.

[edocaster]

Further update: After a total of 17 km (10 with lights off, 7 with lights on), the same phone (this time switched off) has now gone up to 9 bars. I'm happy with that - pretty much as per spec.

However, the phone exhibited some strange charging behaviour, based on what I could see of the charging light. It would light up pretty much at walking pace, but if I accelerated too quickly to beyond 20 km/h the light would disappear. However, if I accelerated slowly, the charging light would remain lit at most speeds (I clocked 40 km/h).

I have no practical way of measuring voltage at the dynamo or USB output while riding (I only tested at 'handspun' speeds before finishing the unit, and it always hit 4.9-5.0v exactly). A total guess would be that output voltage is fine, but somehow the phone doesn't like a sudden increase in available current.

I'm a little worried that when the charging light goes out that leaves the charger in a no-load situation. Is it likely to be OK (I'm unlikely to do more than 50 km/h on this bike)? Or, even though I'm loathe to redo part of the circuit, is there something smart I can add (I might just about have enough room for something parallel across the input (I've heard about something called zener clamps?) or output (a small permanent load?))?

Thanks.