Reflectalite still going: now with LED dynamo bulbs

[SA_SA_SA]

Reflectalite are still going: now with LED dynamo bulbs

http://reflectalite.com/LEDpage.html#anchor78877

The rear one seems a bit(!!) too bright: probably better to stick with a nice approved mudguard or rack LED lamp or 2.

The front one looks interesting for bottle dynamos: needs a filter-backup to avoid flickering with a dyno hub it seems.

(the official bulb list for BS6102/3 was deleted so does that mean its still counts as an approved lamp?).

[Brucey]

Good link!

I reckon for a screw-in tail-light, an NL336R could be the thing. 40 lumens output. If it doesn't work from AC ( it is said to be 'DC' but not polarity sensitive...) maybe a series diode would sort it out.

cheers

[SA_SA_SA]

To avoid flicker a bridge rectifier would be required (NB a smoothing capacitor may make flicker worse: I think because the bulb sees full voltage, attempts full power, empties capacitor then must wait for next half cycle...).

To avoid soldering you could use a heavy duty one (e.g. a KBPC3504 (Maplin.co.uk order-code AR87U) and 1/4in crimp connectors.

40 lumens still seems quite bright for a rear lamp (is equal to filament bulb lumens *4 because the light is already red:

I have dimmed a similar (white terralux led) bulb by a series resistor): it might be easier to dim that official red dynamo bulb by that method (assuming it works).

I wouldn't want to dazzle anyone.

[Brucey]

my (rather crude) experiments re. rear lights show that an 8mm super-bright red LED from maplin (about £1), a single diode (for half-wave rectification) and a suitable series resistor to give ~20mA RMS gives a really good rear light even on an old-style GH6 dynohub, leaving the bulk of the current (in fact all of it on one half-cycle) available for the headlight.

I reckon it is maybe two or three times as bright as a 0.6W incandescent bulb; the 'strobing' is quite attention grabbing at low speeds, and the rear light is quite bright even at a walking pace, especially if the headlight essentially doesn't conduct at low voltages.

Obviously the series resistor must be selected to give the correct current once the generator is under load from the headlight, and the exact value will vary with the headlight properties.

The only caveats to this are;

1) I think it is very important to have a rear light lens with a good diffractive element in it; without this the beam spread rearwards perhaps isn't wide enough.

2) should the headlight fail, the rear light will almost certainly fail instantly unless it is also (say) protected by a zener diode which is in paralell with the LED.

With a full wave rectifier and an LED headlight on an old-style dynohub, I don't find the strobing too bad except at very low speeds. I did try two lights about 14" apart, each firing on the opposite half-cycle, but this gave the kind of illumination at low speeds that could threaten to bring on epilepsy in people who are normally not susceptible....

BTW I am working on a scheme which doesn't 'throw away' 1.4V in a full wave headlight rectifier; I am of a view that 1.8W from an old-style Dynohub is probably enough for road use, provided it is utilised in the most efficient fashion.

cheers

[SA_SA_SA]

Forgot to mention a regulator (reflectalite sell them although two back to back 1.3W/5W zener diodes does the same).

The 5mm maplin leds always seem rated as brighter.

[Brucey]

my suggested zener circuit position will allow the use of a low current rated device, since where it is, the series resistor will always do the bulk of the work.

I don't especially like the idea of simply clipping the maximum (unsmoothed) voltage from the generator since this throws away valuable power. My experiences with this in the past have shown that this can reduce output at medium speeds, even if the light is adequately bright at high speeds.

I am working on a scheme where the front LEDs can use all the current the generator can ever produce without blowing up; this will make for good efficiency at all speeds, and a seriously bright light (which is exactly what you need when going downhill, say,) at speed.

One of the really nice things about modern LEDs is that they remain almost constant in light generation efficiency, even at low currents. Clever LED bulbs will make the most of this (and maybe the reflectalight ones are like this if they adjust quickly enough) but less good lighting circuits end up throwing away more generator power at speed, making no more light.

cheers

[rjb]

I used to cycle to and from work at all times of the day and night as I was a 24/7 shift worker. I used a sidewall dynamo for most of the time and made a voltage regulator using 2 zenner diodes in a back to front configuration across the dynamo output. (ie the 2 posative terminals of the zeners were joined together and the 2 remaining negative leads were joined across the dynamo output. The zeners were rated at 7.5 volts and 1/2 amp. I just assumed that as the voltage rose above the zener reverse voltage it would just act as a short circuit and regulate the peak voltage. During this peiod of commuting i used a single front light (without an internal regulator) with a xenon bulb rated at 2.4 watts (6volt / 0.4 amp) and a seperate led battery rear. I never had a bulb blow over the course of all my commuting and the light seamed quite adequate. BTW I am not an electronics expert, but this worked well for me.

[robc02]

BTW I am working on a scheme which doesn't 'throw away' 1.4V in a full wave headlight rectifier;

Are you thinking about an active bridge rectifier? You might be interested in this.

There is also a lot of good information on candlepower forums - they have a section specifically on bicycle lighting; most of it is about battery lights but there is useful stuff on dynamo lights as well.

I was on there looking for information on an efficient standlight circuit. It seems that the B&M one is quite sophisticated - rather more than a whacking great capacitor and diode(!!) and no one had specific details. (This was some time ago). - I took the easy way out and transferred the whole of the innards of a B&M light into a SA housing.

[Brucey]

If you have 'power to spare' and are using incandescent lamps, simple regulation can be fairly satisfactory. But with low power (and low parasitic drag) generators driving modern LED technology (that produces nice white light efficiently over a wide range of powers), there should no need to throw any power you have generated away.

The active bridge rectifier is interesting; it looks a lot like some circuits that I used to use in a former life; these were high voltage, high power versions of what was essentially a similar thing. The blessed things were always blowing up TBH. I wouldn't rate my chances of designing and building something both efficient and reliable without going through quite a few iterations (and blow-ups) en route....

The secret ingredient in some 'standlight' designs is the sneaky use of a supercapacitor, I am told, but that (for me, anyway...) is a bit further down the line yet.

cheers

[robc02]

The active bridge rectifier is interesting; it looks a lot like some circuits that I used to use in a former life; these were high voltage, high power versions of what was essentially a similar thing. The blessed things were always blowing up TBH. I wouldn't rate my chances of designing and building something both efficient and reliable without going through quite a few iterations (and blow-ups) en route...

This looks more like it - a simple MOSFET bridge! The protection zeners might be worth adding as, under no load conditions, some generators may reach 20V or so - just depends on the generator and switch arrrangement.

The supercapacitor in the B&M standlight is, I think, 1 Farad with about 3V Vmax. As intimated above, it is not just floated across the supply, but actively switched in and out of circuit according to generator voltage and time. There are probably other arrangements to maximise its performance as well - as evidenced by the light (on mine at least) not gradually dimming, it remains fairly bright for a couple of minutes and then goes out abruptly when switched out.

For me, the standlight is a major benefit of modern dynamo lights. I remember seeing an early attempt (commercially available) which used a small rechargeable battery as the backup supply. This was before the advent of supercapacitors, but would be easier to engineer and might be suitable for a DIY solution even today.

[SA_SA_SA]

In the days of filament bulbs, when LEDs were dimmer, I made a 100% efficient rear LED lamp:

by wiring enough paralleled inverse parallel pairs of red LEDs to take 0.5Amp, then wiring in series with the front filament bulb (which acts like a current source ). Delayed lighting of front bulb very slightly but worked.

\Edit current sink changed to current source

[Brucey]

This looks more like it - a simple MOSFET bridge! The protection zeners might be worth adding as, under no load conditions, some generators may reach 20V or so - just depends on the generator and switch arrrangement. .

yes that looks fairly simple and not-very-lossy as well. I don't think the Zeners would be required for an old-style Dynohub but with other generators, yes....

BTW IIRC when I was in Germany a few years ago you could buy rear light fittings that go onto rear mudguards for about £5 that have a standlight built-in. Not sure of the quality or the function tho'.

cheers

[SA_SA_SA]

Reflectalite's website has been down for several days...

[gaz]

.

[nigelnightmare]

I converted an old "Lumotec plus standlight" halogen, last year using one of these:
https://www.ebay.co.uk/itm/NICELITE-LED-BULBS-DYNAMO-HUB-GENERATOR-DYNOHUB-BICYCLE-LIGHTS-REGULATOR-LAMPS/264273458180

Coupled to a fer2002.
It now lights up as soon as you move although the standlight doesn't work (hasn't for years).