dweenk

I have an opportunity to buy a bike with the affore mentioned hub. There is no wiring or light. The wheel seems to be a repalcement. So my question is how do I test the hubs output?
Is a VOM set to AC voltage adequate, and what reading should I expect to see with a vigorous spin of the wheel?

scarlson 

The Old Way was to put a screwdriver across the terminals to short them, give the wheel a vigorous spin, and pull the screwdriver blade a minuscule amount away from one terminal. You should see and/or hear a tiny spark jump the gap. Might take a few tries if it's your first time. Can also do it with a wire if you have one handy. Also, when it's shorted, you will notice more drag and vibration via the axle if you don't have it securely clamped in the dropouts.

It is not harmful to short a dynamo like this. Well, it might heat up if you shorted it and then went on a ride for half an hour.. They are naturally current-limited by the gauge of wire used in the windings, so no damage will occur, unlike when you short a battery (which will try to supply all the amps you can draw from it, right up until the battery explodes).

Digital voltmeters tend to be unreliable for this sort of thing. The voltage and frequency will be varying all over the place as the speed of the wheel changes, and it can be hard for a digital meter to follow. I prefer an analog meter if I'm measuring a dynamo, but in practice I rarely have needed to.

steelbikeguy

I have a digital meter that can measure current, and with that, the most meaningful measurement would be to measure the current out of the dynamo. With a modern dynamo, it should be producing 400mA or more at decent speeds. I know that the SA dynohub produces less power, so maybe 250mA to 300mA?? edit: obviously, this would be a measure of AC current. If the meter is set up to measure DC current, the reading will be zero.

With a literal volt-ohm meter, then you can always check the open-circuit voltage. The problem is that this voltage is mostly proportional to the wheel speed, so the value could be anything. If you can connect a resistor that is close to 12 ohms to the terminals, then an output voltage of 3VAC or more is probably a good sign (but again, I'm not certain exactly what the output of these is).

You could also use the ohm-meter to make sure that there isn't a break in the dynamo's wiring.

The problem with the SA dynohub that I've heard of is demagnetized magnets. This will result in reduced power output. To measure this, you'd need to either measure output current or else connect a known suitable resistor and measure the voltage across the resistor.

The alternative to using a meter would be to short out the dynamo terminals. If it is working properly, it should produce a significant drag when spinning. This is very obvious with the Schmidts that I own. Not sure how obvious it would be with the dynohub, but you should notice a difference between the open-circuit and short-circuit drag.

Steve in Peoria

dweenk

So a simple bulb would be sufficient to verify the dynohubs operation? I can certainly do that with a front or rear light that I have on hand. I have some aligator clips that will make it easy and quick.

steelbikeguy

a simple bulb could work if it is the right bulb. I'm not sure how easy it is to find the right bulb. Basically, it should be rated for the voltage and current that the dynamo produces.
Even with a bulb, it would be good to measure the voltage across the bulb, to know if the dynamo output is up to spec. Of course, you'd have to know the spec... and I haven't looked it up.
Isn't the dynamo output rated at 1.8 watts or something? I don't know if that is 6v at 0.3A or some other combination of voltage and current.

Steve in Peoria

bulgie 

I use a B&M IQ-X on mine, and it gives a lot of light. It's probably not as bright as the same lamp on a modern hub, but without doing a side-by-side comparison, I'm not sure if I can tell the difference. I was pleasantly surprised by how bright it is, very usable, better than a modern dynamo with an incandescent bulb.

I've used Dyno-Hubs with the original correct incandescent, and the light is somewhere in the paltry to pitiful range.

Mark B

steelbikeguy

I've heard favorable comments about using modern LED headlights with dynohubs from other sources too. It also suggests that the dynamo is producing 6v with that load. A further reading of Sheldon's (and John Allen's) info on the dynohub does specifically state that it is intended to be a 6v dynamo.
https://www.sheldonbrown.com/dynohubs.html

I fully agree that dynamos with incandescent bulbs produces a barely tolerable stream of photons.
How did we ever survive? Of course, I do recall augmenting my little Jos block dynamo with a Cateye Micro II light. Between the two of them, I managed to see most potholes and roadkill.

Steve in Peoria

JulesCW 

Surely you're not implying that the depiction of the amount of light cast by vintage dynamos and bulbs in these vintage advertisements is hyperbolic?

steelbikeguy

I'm pretty sure that if I look up the word "advertisement" in a thesaurus, it will list "hyperbolic" as a synonym.

But... to some degree, if you are accustomed to feeble headlights, you'll be happy with something slightly less feeble. Another factor is the ambient light environment. If you are having to deal with the feeble headlights of cars with 6v electrical systems, these vintage lights might not be that bad. By comparison, I see so many modern cars with really bright headlights that are blinding. There's also the tendency of people to turn on their fog lights as a regular driving light. Terrible.

Even with my feeble Jos block dynamo, it wasn't bad on really dark roads. Once your eyes adapt to the dark, not that much light is needed.

Steve in Peoria

tcs

Ah, you, my friend, need a Dynohub™️ Tester!











Manual @ https://www.sturmey-archerheritage.co.../view-1013.pdf

tcs

6V @ 2W until the early 1970s, 1.8W afterwards. There are questions as to whether this was an actual down-rating or just a change in test method.

Sturmey-Archer developed an updated and improved 3W Dynohub in the late 70s/early 80s, called the XAG. This was shown at cycle trade shows in 1982 to positive reviews but - if you know anything about the history of Sturmey, you know where this is headed - they never put it in production!

Fun fact: the GH6 was introduced in 1946, replacing pre-war 12V and 8V models. Despite post-war shortages and supply chain disruptions, by early 1951 the factory was making 12,000 a week!

tcs

Plenty bright during the Blitz!

gilesa

Ass suggested above, there should be 6V of output with a 20 ohm resistor connected, or a little over 300mA short-circuit current. It is hard to get a good measurement while spinning by hand. One method I have used is to use the chuck of a variable-speed electric drill as a roller on the tyre. For comparison, I measured internal resistance at 6.5 ohms and inductance 86-108mH, varying with wheel position.

markk900

You forgot the most important one:

dweenk

Thanks all. I'm going to do some reading of St. Sheldon and decide whether or not to buy the bike.

Road Fan

I agree with a lot of what has been said here. One thing that I think is needed is to emphasize that if you only want to check that the generator is functioning, the VOM needle movement will show that it is really producing voltage, even if the wheel speed is not steady. If you can spin the wheel steadily you should get an AC signal which will look steady on the VOM. Because the meter is mechanical, its response is far slower than for an electronic AC voltmeter. If you look at the signal with an oscilloscope, it should look like a clean sine wave or one with some kind of distortion or spiking at the peaks, due to magnetic hysteresis and saturation. If there's too much weirdness the generator might be no good, or have the magnets out of place in ways that might be hard to understand without a good set of blueprints or CAD.

Road Fan

It's interesting to see your inductance measurement, and that it varies with position. This must be due to the magnetic circuits changing their reluctances as the stator and rotor pole pieces come into and out of alignment. I once designed a driver for an electronic for Ford, and the motor was based expressly on that principle of varying reluctance. Tough to control, but very efficient. I've measured motors in EE labs on special fixtures (not to mention an impedance bridge!), but I really can't attack hub or bottle dynos very well, just at home.

gilesa

To confirm that, the inductance is at a minimum in a stable position of the wheel, where magnet and stator poles should be aligned and the flux path has maximum length. (I had to look up the definition of reluctance!) That was an at-home measurement: wall socket, transformer, resistor, multimeter and Pythagoras. So probably not very accurate.

Greg R

Quick and short, YES. Spinning as fast as you can you should achieve 5-6 volts. Make sure the output nuts are tight whether on a forked connector or just using clips, the nuts complete the circuit, and your readings will be off if they are loose.

You can also test for continuity before the test. The magnets should almost be able to be felt as you gently rotate the wheel, not much but noticeable. That's a good sign right there. Continuity, some magnetism, good signs it's a keeper.

Not many bulbs on the market that are compatible with it's low output. For mine it runs the rear light (cool factor), and I use a separate rechargeable LED for the front.

tcs

Right, incandescent, but the Dynohub™️ lights up modern LED headlamps just fine. I've got mine wired to a B&M OneFive, and Bulgie upthread reported energizing the uber photon cannon IQ-X.

Greg R

Those look like fantastic lights. I seriously considered them months ago. But with the money I already spent on some Bontrager rechargeables, I opted to just keeping the Dynohub running the rear light for now.

tcs

So long as we have a Dynohub™️ thread, let's talk about...threads!

The GH6 (front hub!) Dynohub axle is a 3/8x26tpi. The threads on the terminals are 2BA, a British spark plug thread size.




Now, you're probably asking yourself, 'Self, why is his axle so long?' Well, I'll tell you. Some of the GH6 Dynohubs were built for the old roadster bike front OLD of 90mm. Yeah, that's not long enough to fit in a standard 100mm OLD fork. So I replaced the axle in this NOS Dynohub with a generic one - but, and it's a big but, on a Dynohub axle, one of the cones screws down against a stop. On the generic replacement axle I simulated this with some blue Locktite.