I just got in a commuter bike badged "Pegasus" it appears to have been made to for the European market. It has a Shimano dynamo and came stock with headlight and taillight. The front bearings are frozen and the wires from the dynamo have been pulled out. I found a youtube showing how to disassemble the hub. I wanted to test the lights before I tear into the hub and bearings. How can I simulate the output of the dynamo to test the lights. Can I just hook a 9V or 12V battery to test? Thanks!

I can't say I've ever wired dynamo lights to a battery, but I have two different models of Shimano dynamo hubs on bikes, and both models are 6 volt/3 watt hubs.

The dynohub produces alternating current at 6 volts. A battery *might* illuminate the lamp, but I wouldn't exceed 6 volts. There's a fair amount of circuitry in that LED light, and you don't want to fry it. If you can find a transformer with a 5-6 volt AC output, that would probably be more compatible.
You can pry the grey piece of plastic out of the connector on the hub; the wires appear to have been pulled out, but they are easily replaced. Don't do anything rash.
BTW, the hub is probably fine; there's not a lot to go wrong with it. You could reconnect the wires and test the lights with the hub itself.

Maybe find someone else with a dynamo wheel you can hook up for a test.

The chances the led lights don't work is about ZERO.
Mine have been on 100% of the time for 33,000 miles/ 13 years with 2 tours. The first one had hot and cold and wet weather.
But hey, first wash your fricking bike.

What sort of lights? Some have a built-in regulator to stop the dynamo from burning out the light at high speed. Without a regulator I wouldn't risk 12 volts or 9 volts, but an old USB charger runs at 5 volts which would be good if you can find the right wires to use. Note that some LED lamps only run with the right polarity, so if it doesn't appear to work reverse the wires.

One aspect of these hubs is the vastly different feel when spinning the axle by your fingers. I can't remember a modern "dyno" hub that didn't feel like the bearings were roached, the alternator induces rotational spots of "grab". IIRC The SON US importer (Peter White Cycles) has a number of service warnings about servicing their hubs, almost as if many will mistake the alternator axle rotation feel for bad bearings. Andy

I don't know diddly about Dynamo/LED bike lights, but I used to build an oceanographic instrument that used an LED as a light source.


Besides voltage, CURRENT can also kill an LED. Hooking up a non current restricted 6V supply might just fry things.
EDIT added- a 2k resistor in series w/ the +6V would limit current to 3 milliamps.
To limit a 6V supply to 3W, you'd need a 2 OHM resistor.
"...the minimum voltage needed to light it, varying by color from about 1.6V for Red to 3.6V for Blue/White LEDs, with lower voltages for infrared and higher for blue/UV..."

LEDiode. It only conducts one way because it's a DIODE. Only 1/2 of your AC sine wave would do anything.

I would presume the light "assembly" probably has a full wave bridge rectifier to convert the AC to DC. Possibly some type of current regulation if not simple reliance you aren't going to over speed the hub?

Anyone with enough morbid curiosity to see what the inside of one of these hubs looks like can find images HERE. This was a dissection I did about 15 years ago after I was told this particular dynohub was "not serviceable". It was a real learning process.
The OP's hub probably differs from the early Dahon "Joule" hub I resurrected, but many of these hubs look like they may have "common ancestors".
To Andy's point, the hub's axle will will have a distinctly "notchy" feel as it's rotated, due to the magnets passing the poles of the stator coils. This, plus the rapid spin-down of the wheel (also from the magnets) makes evaluation of the bearing conditions in these hubs difficult.

Remove the wire block (pull firmly up with thumb/forefinger on the black, textured block below the grey piece). Spin the wheel, and hold a key to the exposed metal tabs--if it sparks, there is nothing wrong with the hub, and you can safely re-wire the hub (instructions can be found under "dealer manuals" at si.shimano.com after you put the model of the hub (DH-XXXX) in the search bar.

This is good information. However... you don't get much of a spark at 6 volts/3 watts unless the key position is just right. It would be easier and more predictable to get a 6-volt flashlight bulb and wire it to the hub.

I couldn't find a spare connector to do a pictorial, but this image should be clear.
The connector for my early "Joule" hub is a gray piece of plastic that holds the wires. Once the wires are installed, the assembly simply plugs into the receptacle on the hub.
On the Shimano hub connectors I found on the web, there appears to be a separate plastic sleeve that fits over the piece with the wires. (FWIW, Shutter Precision dynohubs also have a similar 2-piece connector.) This assembly slides over a "tab" sticking out of the hub; the tab bears the two connector strips.
It doesn't matter which style of connector you have; wire installation is the same for both.
First, strip enough insulation off the wires to have about 1/4" or so of bare wire extending from the connector. The insulated wire should be fully inserted in the gray piece.
Fold the wire over the plastic piece as seen in the image below. The wire strands should be well twisted to prevent stray strands from causing a short. I like to "tin" the wire ends with solder; this prevents shorts as well as making frequent removal and re-installation easier (if you need to do this).
Insert the assembly into the receptacle on the hub, or if you have the 2-piece connector, snap the two pieces together and slide them over the tab on the hub. I like to coat the inside of the connector with silicone grease, which makes the connection more resistant to corrosion (helpful if you ride on salted streets).


Trim the wires so they stop just short of the lip on the piece at the top (this one is black; usually they're gray). This is a 2-piece connector.

I suppose the thing that fails most commonly is the wiring between the hub and the light. I would try testing or replacing that first.

Yes - "dynamo" hubs are actualy alternators - dynamos produce "direct current" (voltage actually, technically), but these hubs produce alternating voltage.
"Dynamo" has become a generic term for the "thingy that produces voltage."

And for the LEDs, yes, strong current can fry them.

Shimano sells a "surge protector" which I am not sure how exactly it works (haven't diassembled it) - but I suppose it does effectivelly limit the current running through a light:

https://si.shimano.com/de/pdfs/si/2Z...0A-000-GER.pdf

Current is a product of Voltage and Resistance (or Impedance). High current doesn't just "happen" unless resistance/impedance decreases or voltage increases.
The circuitry in the LED light doesn't change, so the only realistic possibility is over-voltage from the dynohub.
There is an interesting phenomenon with alternators: as they spin faster, there is an increase in the frequency of the current produced. Since the current is flowing through coils of wire with iron cores, as frequency increase the impedance increases... which limits the current to some extent. This may be enough voltage regulation to prevent LEDs from being fried by high speeds.

But LEDs don't obey Ohm's law because they're not simple resistors - once past the voltage threshold, conductance increases rapidly but voltage drop is constant. That is why in a simple circuit you add a series resistor to stop them blowing up, for limiting current, not voltage. I won't pretend to understand the operation of alternators, I remember left hand rule vs. right hand rule but that current source vs voltage source and back EMF stuff seems a bit magical to me - and what's with the potentially huge open circuit voltage?
Anyway, seems like you want a full wave rectifier with smoothing and voltage regulation and some form of current regulation to make a dynamo (actually permanent magnet alternator) a suitable power source for LEDs.

I'm relying on electronics I learned 60 years ago, so I should probably just shut up. Suffice to say that there's a lot of "driver" circuitry in most LED lights, and this includes whatever is necessary to protect the LED itself.

Interesting thing about the dynohub's output. Since it's AC, and the frequency varies with road speed, a microprocessor monitoring the frequency can be used to activate a brake light.
Busch & Mueller makes this one; I have one on my road bike and it actually works. No physical connection to the brakes.

Maybe add a cadence sensor so it knows that you're slowing down while not pedaling, rather than struggling up a hill. An added accelerometer could tell the difference between hard and gentle braking, to show more/brighter/flashing light. Add a death grip sensor to the handlebar, a pucker sensor to the saddle ...

Hahaha! I don't know what the design parameters are, but it does brighten up during "normal" braking.
The "pucker" sensor is a great idea!

Yup - it's awesome, use it too.

Haven't disassembled it, but my guess would be that it reacts to a frequency drop.

This is what I've been given to understand. Whatever the mechanism is, it works.