Run rear light from same dynamo as front

[steveo_mcg]

What do I need to add to run the rear light off the same hub dynamo as the front? The front is 3 xpg emitters the rear is a single high power Cree red emitter. Do I just need a second bridge rectifier or do I need a resistor to drop the voltage on the circuit from the 7v or so from the front?

 

[Steve K]

Re: Run rear light off same dynamo as front

The nice thing about home built bike lights is that you can design them to your preferences. The disadvantage of that is that you have to know enough about electronics to take care of the problems that might pop up.

The two primary options that you have are whether you want to wire the taillight in series with the headlight or in parallel. The commercial bike dynamo lights use the parallel arrangement, which has the advantage of not losing both lights if one of the wires opens up. With a parallel arrangement, the taillight will receive AC power, so there will need to be someway to use this power. A bridge rectifier would be the most common method. You'll also need to regulate the power to the red emitter too. A resistor might be the easiest technique. When choosing the resistor value, you'll want to remember that whatever current you put through the taillight is current that is being taken away from the headlight. Also consider that since your headlight will have about 9v across it, you'll have to drop a lot of voltage across the resistor in order to limit the current to the red Cree. You might consider using two red Crees in series in order to make the best use of the power being delivered to the taillight.

If you choose to wire the taillight in series with the headlight, the headlight and taillight will receive the same amount of current, so the red Cree will need to be able to handle 500mA. The taillight will still be getting AC power, so it will need to be able to pass current in both directions. A bridge rectifier would work, but will add more voltage drop to the headlight/taillight combination. An alternate method would be to wire a plain 1N4001 diode across the red Cree, anode to cathode, so the LED will carry the positive dynamo current and the plain diode will carry the negative current. Or use another Cree LED instead of the plain diode.... both will work.

The downside of wiring the headlight and taillight in series is that you've increased the voltage that the dynamo must produce before current starts to flow. This means you'll have to be riding faster before the lights start to work.

This isn't a short answer, but is a quick glimpse into the options that you have.

 

[steveo_mcg]

Re: Run rear light off same dynamo as front

The nice thing about home built bike lights is that you can design them to your preferences. The disadvantage of that is that you have to know enough about electronics to take care of the problems that might pop up.

Or have some one kind enough to share their knowledge

Parallel I've pretty much got my head around. I've been meaning to change my head light to a twin xml as my current triple uses copper end caps for the xpg's and is very heavy so that would drop the amount of voltage in the circuit to about 6v with two rear emitters that wouldn't be that much loss over the resistor. In parallel the two circuits are effectively separate so the rear must be able to handle enough voltage to make the front turn on, yes?

I'm not sure I follow on the serise arrangement. If the voltage is rectified in the front light circuit how can the rear receive ac? Or do you pass the current through the rear effectively unrectified, with as you describe a pair of diodes opposite each other, and then on to the bridge for the front? I think I like this arrangement but I need to work out the physical layout on the bike, its a little more complicated. Also doesn't the red emitter get too much voltage when the front is up to power? And how could one get some DC to charge a supercap?

 

[steveo_mcg]

Re: Run rear light off same dynamo as front

Ok, I've slept on it and given it some thought. Pragmatism has won out over efficiency. For the time being I need a rear light and won't have much time to redesign the whole assembly so...

For the rear I'm going to go with Steve's suggestion of two 500ma red cree's and for the sake of getting finished a big resistor. I realise i'm wasting a lot of power given a single 500ma red would be plenty plus the 3w or so lost to the resistor but needs must. So could some one check my workings before I blow the lights up...

At full power the front array draws about 9.5v (400ma dynamo) the rear with two emitters in series draws 4.2v so R=V/I : R= (9.5-4.2) / .4 = 15 ohms?

That should give me enough power for the front and keep the rear simple enough I can get it running this weekend. In the next few months when I change my front to a twin xml arrangement the overall voltage in the system will drop by about 4v and I'll only need to drop a volt or so in the rear so I can just change the resistor in the rear and all is good.

The series arrangement still intrigues me as it seems much more efficient but I just can't picture it, I don't suppose some one has a circuit diagram I can mull over?


URRRGH, i've just reread Steve's post and now I realise my parallel set up won't work since the front and rear will be sharing the 500ma not getting 500ma each. So series it must be.



Edit the third: Ok I think i've got my head round the series thing, would this circuit work? If so why doesn't the voltage smoke LED 1?

 

[yellow]

anything other than some 5 mm led for the back --> hopefully police catches the person.



taillight to be wired in parallel to main light:

 

[steveo_mcg]

Re: Run rear light off same dynamo as front

anything other than some 5 mm led for the back --> hopefully police catches the person.

Really? I don't see any 5mm leds in this picture. It might be different where you live but here even 5mm blinkies on epileptic seizure inducing modes are completely lost in the general volume of light from the traffic. I ride on the roads with the car, bus and truck traffic and need to be at least as conspicuous as a side light more so if possible, a reflector is more effective that little 5mm emitters.

Even a pair of high power emitter running at an amp is less than a pair of brake lights so I'm not sure what the problem is tbh...


Thanks for the diagram though.

 

[Steve K]

Re: Run rear light off same dynamo as front

....URRRGH, i've just reread Steve's post and now I realise my parallel set up won't work since the front and rear will be sharing the 500ma not getting 500ma each. So series it must be.

Parallel can still work, but you'll have to share the 500mA between front and rear lights. The usual arrangement in dynamo systems is to allocate 400mA to the headlight and 100mA to the taillight. The proper selection of series resistor at the taillight should do the job. The only question is whether 100mA produces as much light as is desired.

Edit the third: Ok I think i've got my head round the series thing, would this circuit work? If so why doesn't the voltage smoke LED 1?

The dynamo has a large internal impedance that will limit current to about 500mA, but the voltage can be rather large depending on the load it is driving. As long as the LED can handle the 500mA, you are fine.
My only question is why there are two red LEDs wired in series. This will be a pretty large overall voltage drop, which will require higher speeds in order to start lighting up the LEDs. My inclination would be to just use two red LEDs wired back to back as the taillight. ....well... my actual practice has been to use ten 5mm red leds wired in parallel instead of a red Cree. The voltage drop is usually 1.7v or so, instead of approx. 3.0v.

 

[steveo_mcg]

Re: Run rear light off same dynamo as front

I think the problem with parallel is my cheap hub only generates 400 ma and losing 100ma might lose too much brightness up front on unlight roads.

I think one should be sufficient if its getting 400 ma. Do I need a blocking diode on the reverse of the diode to stop it getting damaged on the "negative" part of the wave? Ok i'm off to do some experimenting I find it much easier to get this stuff straight in my head with real parts rather than theoretical diagrams. Thanks for the help I may be back :-D.

 

[Steve K]

Re: Run rear light off same dynamo as front

When the taillight is wired in parallel with the headlight, you won't get any voltages above 9vac or so across the taillight. 9v is enough to kill a lot of modern LEDs, though, so you'll want some protection from the negative portion of the AC waveform. Besides, the headlight might have some sort of open-circuit failure, which would allow the dynamo voltage to increase quite a bit.

 

[steveo_mcg]

Re: Run rear light off same dynamo as front

Reverse polarity protection is confusing me... Should the band end of the diode be connected to the negative or the positive marked terminals of the emitter?

Also I tried to connect a super cap across the Cree as a stand light, should this work? It wasn't but I'm prepared to say I've done something wrong.

 

[Steve K]

Re: Run rear light off same dynamo as front

This shows how one of my designs is set up:

The diode D1 protects the LEDs to the right from the negative portions of the dynamo's AC output voltage.

 

[steveo_mcg]

Re: Run rear light off same dynamo as front

Am I reading this correctly? are the emitters all laid out parallel? Sorry I'm not great with proper circuit diagrams.

I gave up with the protection diode and wired up the two red emitters parallel and its working very well.

But I cannot work out how to get a stand light, I thought I'd have sufficient dc across one of the emitters but it appears not and my rectified dc in my head light is in a sealed box.

 

[Steve K]

Re: Run rear light off same dynamo as front

Am I reading this correctly? are the emitters all laid out parallel? Sorry I'm not great with proper circuit diagrams.

I should mention that the taillight is built with 10 red 5mm LEDs in parallel. For big LEDs, you would just use a single LED.
One nice thing about using an array of 5mm LEDs is that it makes the light source bigger. This reduces the glare (the intensity at the light source). The Vf of the 5mm LEDs is close to 1.8v, IIRC, versus the larger Vf for some of the large high brightness red LEDs. I suspect that the 5mm LEDs aren't as efficient in turning electrical power into photons, though.

I gave up with the protection diode and wired up the two red emitters parallel and its working very well.

But I cannot work out how to get a stand light, I thought I'd have sufficient dc across one of the emitters but it appears not and my rectified dc in my head light is in a sealed box.

Two LEDs wired in parallel but opposite (the anode of one hooked up to the cathode of the other) works but has the disadvantage of using an extra LED and optic.

There was a long thread about standlights a couple of years ago (+/- 5 years). Making a standlight that is supposed to power an AC light like this is not as simple as slapping a supercap across the LED. You could use the standlight circuit that I have in my schematic, but it obviously is more complex and adds a lot of parts.

 

[steveo_mcg]

Re: Run rear light off same dynamo as front

Cheers Steve, I can see the advantage of then 5mm the spread would be much easier to achive than with cheap dx optics but I'd need to design a new body, the one I have only needs modified.

I did some reading and I think it was a comment of yours about protecting the cap from the negative portion so I thought I'd try this. It's probably not that efficient but it is simple .

 

[steveo_mcg]

Re: Run rear light off same dynamo as front

In case any one comes across this thread in the future...

The above works fine, 20f supercap takes a few minutes to charge from empty and last more than long enough as a stand light. The caps are probably in the middle of the internal resistance range, they're not very high so they discharge quickly enough to give plenty of light just as I stop but they're resistance is high enough to not need a resistor to limit the discharge current. In real world use, they don't take that long to charge in the morning because there is still some juice in the capacitor and when travelling at cruising speed the flicker on the non capacitor levelled is not noticeable.

A better solution would probably be one emitter and an extra diode to boost the voltage stored in the capacitor but I didn't have time to continue experimenting and light has been in daily use for a week or so now. My other plan to swap the front for 3 XR-e to 2 XM-l would also improve efficiency and reduce flicker on the whole array.


Steve K, thanks a lot for your assistance.

 

[Steve K]

Re: Run rear light off same dynamo as front

Out of curiosity, what cap are you using? i.e. manufacturer, model, etc. I haven't shopped for supercaps in a long time, and the supercap that I did buy just didn't impress me as a standlight cap.

Regarding the supercap not taking much time to charge up: I think that most folks run the calculations without remembering that the supercap really can't discharge much below the LED's forward voltage (Vf). For a typical modern white LED, the Vf will be within a few tenths of a volt of the nominal Vf as current ranges from the rated current down to a few mA. As such, it really only has to be charged up by a half volt or so, which shouldn't be a huge current hog.

 

[steveo_mcg]

Re: Run rear light off same dynamo as front

I'll check when I got home but they're ebay originals so getting the same ones is probably a bit of lottery.

When I was first bench testing I thought there must be an issue with the circuit because nothing happened when I hand spun the wheel after a moment I realised the cap would be completely empty and hand spinning doesn't generate much power so I juiced it with a pair of batteries and it all worked fine.

 

[steveo_mcg]

Re: Run rear light off same dynamo as front

Steve, they are Yec brand 10f 2.7v caps (ebay item 261113249440), I figured for the rear two in parallel would give me 20f and the voltage would be with in tolerances and upfront I could run them in series to up the max voltage, bit more flexible than a single 20f and higher capacity than the 5.5v ones.

datasheet

 

[Steve K]

Re: Run rear light off same dynamo as front

10mm diameter x 30mm length.. not too bad. $17USD? a bit pricey for my tastes, which is probably why I keep using a AA nicad cell for my standlights. Sort of a false economy, I suppose, since I add so many hours of labor to put together my little boost converter. I suppose it's better to not think too much about the cost of the personal time when it comes to hobbies like this, though.

 

[Bobblehat]

Re: Run rear light off same dynamo as front

Steve, they are Yec brand 10f 2.7v caps (ebay item 261113249440), I figured for the rear two in series would give me 20f and the voltage would be with in tolerances and upfront I could run them in parallel to up the max voltage, bit more flexible than a single 20f and higher capacity than the 5.5v ones.

datasheet

Short note on above: I think 2 in series would give you a 5F 5.4V equivalent, and 2 in parallel would give you a 20F 2.7V equivalent. You'd need to take steps to make sure the caps were not getting more than their V rating in either case.