3 to 9W VPG … variable power generator

[NATURAL ANKLING]

Hi,
Field, stator, rotor, armature are sometimes used in loose terms, this might clarify that, if you are so inclined........or not
http://en.wikipedia.org/wiki/Field_coil

[ConRAD]

... for simplicity a multi take off of field coils, either several windings sharing the same pole or just independant poles with separate winding, mean that independant take offs for lights and ancilleries, reducing the need for extra components for the lights...

... for simplicity:

1. Take-off windings must be stationary, no brushes, no slip rings are allowed

2. No multi take-off field coils, no several windings sharing the same pole, no indipendant poles with separate windings shall be used: AC output must be unique, i.e. 2-off wires only. Then AC output shall be split one way to power the lights the other way to feed an AC/DC converter.
At the input of AC/DC converter, available power, on demand, shall have to be not less than 6 to 9W.

3. Shouldn't entire full power be strictly necessary, a "smart" provision mechanism to reduce permanent magnet back-dragging shall be activated

[axel_knutt]

… why not a 3 to 9W variable power generator?

I think what you’re asking is why not have a dynamo that will drive any lamp between 3W and 9W. The short answer is because of the simple method a bike dynamo uses to keep the output relatively constant as the speed of the bike varies. For a longer answer we need to start from square one.

When you move a conductor through a magnetic field it generates a voltage which is proportional to the velocity, so the output voltage of any generator is proportional to the speed at which it rotates (as is the frequency). If the generator is going in somewhere like a power station where the turbines turn at a constant speed that’s not a problem, but the speed of a bike varies by 10:1 or even more, so that leaves you with a corresponding variation in supply voltage unless you do something about it. What the manufacturer is looking for is a cheap and simple way of keeping the output to a lamp constant, or near enough constant to prevent the filament blowing.

The way it’s done is using an inductive source impedance. The reactance of an inductor is proportional to the frequency; so as the frequency is proportional to the speed, you will have a source impedance which is nearly proportional to the speed once you’re going fast enough for the reactance to be significantly larger than the sum of the source and load resistances. Current is voltage divided by impedance, so since the voltage and the impedance are both proportional to speed you end up with a current which remains independent of speed.

At this point we need a little digression to understand the difference between constant current and constant voltage sources. When electronics engineers use these two terms, unless explicitly stated otherwise the word constant refers to variations with respect to load resistance and not speed or any other parameter. As we’ve seen, we’ve ended up with a supply which has a voltage and current which are both fairly constant with respect to speed, but this is a constant current supply, because it is the current and not the voltage which remains the same as the load changes.

A generator approximates a constant current when the load is much lower than the source impedance because total circuit impedance is relatively independent of load changes and thus current is relatively constant. However if the load impedance is much higher than the source, the voltage lost across the source impedance is only a small fraction of the total, and so you now have a constant voltage source. In our case, the manufacturers have designed the source impedance to be much higher than the load, so we have a constant current generator and not a constant voltage. The load voltage is only constant so long as the load resistance remains constant.

Fine so far, but power is both V^2/R and I^2R. Bulbs are designed to run off a constant voltage supply, not constant current, so the resistance of a 9W bulb will be a third of that of a 3W bulb that runs off the same voltage. If you now connect it to a supply that delivers a constant current that’s correct for the 3W bulb you will end up with a 9W bulb that’s only actually dissipating 1W. Alternatively, a 1W bulb would have the right resistance to produce 9W, but it will promptly blow because it’s dissipating nine times its rated power. An 18V 9W bulb would have the correct resistance, but then you will have to cycle about three times faster before the output of your dynamo stabilises enough to give you a relatively constant brightness.

So, the reason that the same dynamo can’t run different size bulbs is that the source impedance has to be tailored to a specific bulb size so as to take advantage of a cheap and simple way of regulating the output. Whatever the size of the dynamo you still come back to the same problem of obtaining a constant output. A generator with a low impedance output and a solid state regulator will keep the voltage stable as the speed varies, but how many are going to pay for it? Something like the B&M E-Werk costs around ten times the price of a cheap dynamo, so it looks as if they’re having trouble finding enough demand to get the economies of scale.

[ConRAD]

… why not a 3 to 9W variable power generator?

Definitely clear, thanks a lot.
I’m wondering though … why at this point dynamos are universally rated 6V-3W and not 0.5A-3W ??
In modern dynamo/frontlight sets (e.g. SON/Edelux) dynamo output voltage at no-load conditions ranges in the between “0V” and nearly “50V” depending of course on speed. With light switched-on voltage is limited to 7.6V approx by a voltage limiter built-in the frontlight itself. So since we are now not talking anymore of filament bulbs but rather LED-electronic controlled devices, it seems that the dynamos originally designed as constant current sources actually are now somehow “forced” to behave as a constant voltage supply (indeed, it’s inside the headlight that electronics drive the LED at continuous/pulsating constant current).

Now, going back to the initial question: … why not a 3 to 9W variable power generator?

Well, my query originated from the fact that all bicycle dynamos on the market are basically designed all the same: rotating permanent magnets, stationary armature coils, that’s it.
The problem is that no-load dragging of a dynamo rated 9W is of course much more than a dynamo rated 3W, simply because of its stronger magnetic circuit.
It appears that some manufacturers in the recent past (e.g. Supernova with Infinity 8 dynamo) tryed to overcome the no-load dragging problem introducing a sophisticated unclutching mechanism to somehow disengage the magnetic circuit when this one not in use, i.e. when ligths are switched-off.
I’m not sure whether this solution is still into production or if it has abbandoned ... because perhaps too complicate.

In short: since I’m still convinced that cycle dynamos will be based, I guess for a long time yet, on permanent magnets, why not modulate their output power in such a way that when I need 3W I take 3W … but when I need 9W I’ll be able to take 9W ??

[NATURAL ANKLING]

Hi,
Its in that word "Variable" which mechanicly is difficult hence the two speed Dynamo described earlier, but it will still top out at same power when the mean rpm and mean max output is reached.

A brushless alternator could be used but I fear the cost of this micro unit which could do everything you require (still needs box of tricks) would be exorbitant.

[ConRAD]

...a brushless alternator could be used but I fear the cost of this micro unit which could do everything you require (still needs box of tricks) would be exorbitant.

... exorbitant ?? ... no brushes, no slip rings, no magnets, absolute no dragging when not in use, just a few reliable diodes, that's it !!!
IMO definitely cheaper than most of dynamos on the market today !!
"BRUSHLESS" ... that's the key word to "3W to 9W variable power generator"

[NATURAL ANKLING]

Hi,
Wow so Its on the market in minature form, can we buy it

I think when a manufacturer gets an Idea that its a mean green unit we might see it on bikes, until then we are dreaming unless you know somthing
http://en.wikipedia.org/wiki/Electric_generator

"Some of the smallest generators commonly found power bicycle lights. Called a bottle dynamo these tend to be 0.5 ampere, permanent-magnet alternators supplying 3-6 W at 6 V or 12 V. Being powered by the rider, efficiency is at a premium, so these may incorporate rare-earth magnets and are designed and manufactured with great precision. The maximum efficiency is around 80% for the best of these generators—60% is more typical—due in part to the rolling friction at the tire–generator interface, imperfect alignment, the small size of the generator, and bearing losses. Cheaper designs tend to be less efficient. Due to the use of permanent magnets, efficiency falls at high speeds because the magnetic field strength cannot be controlled in any way. Hub dynamos remedy many of these flaws since they are internal to the bicycle hub and do not require an interface between the generator and tire. The increasing use of LED lights, more efficient than incandescent bulbs, reduces the power needed for cycle lighting."

http://en.wikipedia.org/wiki/Micropower

"Piezoelectric nanofibers in clothing could generate enough electricity from the wearer's body movements to power small electronic devices such as iPods or some of the electronic equipment used by soldiers on the battlefield, based on research by UC Berkeley Professor Liwei Lin and his team. One million such fibers could power an iPod, and would be altogether as large as a grain of sand. Researchers at Stanford University are developing "eTextiles," which are batteries made of fabric, which might serve to store power generated by such technology"

[ConRAD]

As an alternative to 3 to 9W variable power and just in an emergency situation, is anyone using a supplementary bottle side dynamo in addition to the hub dynamo ?? I was wondering about the 12V-6W B&M bottle side dynamo, is there anybody using it ?

[Brucey]

In short: since I’m still convinced that cycle dynamos will be based, I guess for a long time yet, on permanent magnets, why not modulate their output power in such a way that when I need 3W I take 3W … but when I need 9W I’ll be able to take 9W ??

well you can do that already, at speed; you just need to alter the impedance of the load that you attach to it.

Simple example; most hub generators will drive two 6V 3W lights in series when the speed is over ~16kph. You can drag more power out than that with the right circuitry.

BTW if you want the generator to be efficient on-load, the magnetic circuit needs to be efficient. If you make an efficient magnetic circuit, the no-load drag goes down too.

The best hub generators presently slow you by ~0.1kph, lights off. Even a cheap shimano hub generator will only slow you by ~0.2kph, lights off.

You hanker after a generator which is wonderful and amazing, but I don't think you really appreciate how wonderful and amazing the ones we have at present really are.

cheers

[ConRAD]

Unfortunately, according to "official" data supplied by SON and assuming an efficiency of 65%, you'll never get more than 3.90W from a SON28 at 30 km/h.



... so, again:
as an alternative to 3 to 9W variable power and just in an emergency situation, is anyone using a supplementary bottle side dynamo in addition to the hub dynamo ?? I was wondering about the 12V-6W B&M bottle side dynamo, is there anybody using it ?

[Brucey]

You are looking at a chart of total loss power vs speed when the generator is powering a single light there, and then assuming a given efficiency and therefore output power. There is no need to look at that chart and it is misleading anyway because the generator is a lot less than 65% efficient at most speeds. That data is taken using the 'regulation load'. The generator is designed not to greatly exceed 3W when driving into this load; that is the German regulation. Nearly every generator made to this specification will produce a lot more power than this (and create more drag...) under other conditions.

If you alter the load, you alter the power output.

Try it for yourself and see.

As I said before there are loads of people out there running two headlights in series, giving 6W output at speeds over a certain level.

cheers

[edocaster]

The SON data is probably just based on a 12 Ohm 'standard light' load.

You can get significantly more if you use a higher load: http://pilom.com/BicycleElectronics/DynamoCircuits.htm - you can see a chart there with 11W at 30km/h. I've personally experienced 7.9W at about 25km/h with a Shimano hub (using 5 LEDs).

I haven't read this whole thread, but I think most current dynamo hubs can get 9W at reachable speeds.

(Edit: Ooops, looks like Brucey beat me to it )

[andrew_s]

A generator with a low impedance output and a solid state regulator will keep the voltage stable as the speed varies, but how many are going to pay for it? Something like the B&M E-Werk costs around ten times the price of a cheap dynamo, so it looks as if they’re having trouble finding enough demand to get the economies of scale.

The Lightspin sidewall dynamo was like this (I believe), but is no longer available except perhaps on eBay.
It came out at much the same time as the SON (1988, ish), cost nearly as much (£90 vs £105 for the SON, iirc), and had problems with the on/off catch so many users ended up holding it off the wheel with a bungie.
You could spin the roller with your fingers and it would spin for 30 sec. If you used two lights they had to be wired in parallel (rather than series like a regular dynamo) (but running two lights gave problems with roller slip every time it rained).
Everyone who wanted something better than an Axa bought SONs instead, so they ended up going out of production.

[jb]

The 'Lightspin' eh?
I have one - if ever there was a ship spoiled for a ha-path O tar then this was it. When it worked it was great.

[ConRAD]

...if you alter the load, you alter the power output ...

... definitely ok, but:

the truth is that at 20 km/h I was getting less than 3W, and considering that speeding-up to 30-35km/h just to get one or two additional “usable” watts it would require an estimated 100W supplementary cycling power … well, for me, it was simply a nonsense !!!

So far I confirm: I’m still thinking of a generator capable to produce a 3 to 9W variable/usable power at the speed of 20 km/h !!!

All the details HERE