Hi everyone, I've run into a bit of a strange issue with my lights over the past couple of months. For background, I've been running on this setup for the past 3 years roughly, it's a Shimano DH-UR705 hub, wired up to a B+M IQ-X, with a toplight line small taillight. I had the headlight replaced under warranty in the summer as it had failed and stopped illuminating.

The problem: Every time I start a ride, the lighting system takes about 30 seconds of riding at 20-30kmph before it starts working. After that though it works completely fine for the rest of the ride. I've checked all my wiring, connections, and have re-wired the connection to the hub to ensure everything has a solid connection. Before this, the lights would turn on immediately as soon as I started riding. Additionally, on longer rides in daylight when I sometimes turn off the lights using the button on the headlight, they will spontaneously turn on again for no apparent reason.

I haven't been able to find any other threads anywhere on the internet that quite describe this issue, I was wondering if anyone has any insight on what might be the problem here? Is it time to get a new hub? I've ruled out the light as the problem just due to the fact that it's an almost brand new unit.

Thanks!

Ideally, someone would recognize the symptoms and tell you what the problem is.
That someone isn't me.

In the meantime, I'd guess that the majority of bike light problems are related to the wiring and connectors. You imply that you've eliminated that possibility, however.

You could disconnect the tail light, just to ensure that it's not the source of the problem.

The headlight is the most complex item in the system, which suggests that it has the most ways to fail. The classic way to find out if it is bad is to swap in a known good one. That might be the hardest part of this technique. How well does your local bike shop like you? You could just buy a 6V, 0.5A bulb to wire in place of the headlight to just see when/if it lights up.

Another option is to run the headlight from a 6V, 0.5A power supply and see how it does. The complication here is that the light might need to sense the AC voltage to work correctly. An AC transformer with a 6V output might do the job, but the light might expect a current limited power source, so this has risks too.

That's the limit of my good and bad ideas. Maybe someone else has better info than I do.

Steve in Peoria

I have used a Li Ion battery to test dyno powered lights. At roughly 3.8 to 4.1 volts DC, their voltage was enough to light up the lights without any danger of damage. At that low voltage, they are not really bright, but bright enough to see.

For testing I used a phone battery from the era when phone companies allowed consumers to change their own batteries. I also have removable Li Ion camera batteries that I have used for testing.

If the lights work when testing this way, I would try checking the contacts again, perhaps cleaning off the contacts with rubbing alcohol.

If you have access to another dyno powered headlamp, you could swap that out to see if that solves it.

I think all Shimano dynohubs are grounded to the fork, I know that some are. Is there any possibility that some of your wiring has an intermittent ground to the fork or frame?

I can't think of anything else that SteelBikeGuy did not already address. He had a very good comprehensive list. Good luck figuring this out.

There's a capacitor in there that takes about that long to charge up if you haven't ridden for a while. Is the stand light working OK?

x2

might have a swoll cap?

The hub will be fine. The light can't do this as it's against the StVZO regulations (to not give light out whilst initially charging) and I would say there's something wrong with it.

Suspect it's another warranty job, sorry.