LED Signal Speed

Either:

1) get a resistor which draws the same as the bulb and wire it in parallel with each LED lamp (sort of defeats the energy saving property of LEDs).

2) get a different flasher module which isn't sensitive to load. These are available for cars/trucks towing trailers (where the lamp configuration changes) but I don't know if they'll fit motorcycles.

3) travel fast enough so that the relativistic time contraction will cause time on the moving body to appear to pass more slowly to a stationary observer. To them, the flasher will appear to be operating at the correct rate. ;-)

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
RAM disk is *not* an installation procedure.

Here's a novel idea, run the positive signal wire to a FlipFlop in toggle mode via a 1 or 2 K resistor, then run either of the FlipFlop's outputs to a transistor via a resistor of suitable value to fully turn it on. You then can wire your leds to the emitter/collector circuit in series with your mains power and an appropriate resistor for correct current draw. This will exactly half the flash rate. Go back to your college notes and use ohm's law etc, if your not sure about a few things. Once you've mounted it all on a 'Vero Board', including all of the wires hanging out being the same gauge as the bike's, you could mould it all in a small oblong shape with an apoxy resin or silicone, for weather proofing. (Good idea to make sure it works perfectly and all soldering is robust and clear from shorts first) Then crimp the correct connectors to your wires and whack it in ya bike mate. I'm no expert either (which will be plainly obvious to many) so I cant recommend which tranny to use or weather a TTL or CMOS FlipFlop should be used, but the concept is certainly viable and there's bugger all cash to part with if you use some of your resourceful side.

All the best, Mark Kelepouris

................................. and now for the attacks and grilling?

I like Paul's idea. But if you want to build it yourself, go here:

Build the circuit in figure 5, substituting a 100K potentiometer for R1 and R2, with the pot's center pin connected to pin 7 of the 555. Where the diagram shows one LED and a 220 resistor connected between pin 3 and ground, you will put 6 Leds in series with a 150 ohm resistor. To run 12 LEDs, make another string of

6 LEDS in series with a 150 ohm resistor, and connect it from pin 3 to ground, just like the first string. You omit the part that shows an LED in series with a 220 resistor from pin 3 to (+).

Your circuit will run from the 12 V motorcycle battery. Add some transient protection this way: connect a 10 ohm, 10 watt resistor to +12. Connect the other end of the resistor to pin 8 of the 555 and to the banded end of a 1.5KE15A-T transorb. The other end of the transorb is connected to ground. The transorb is available from Digikey - part #1.5KE15ADICT-ND Adjust the pot for the desired blink rate.

Ed

Your right, Paul's idea is very likable, especially for it's simplicity. He's resistor (1) idea might not have the energy saving side of LED's properly exploited, but the other benefits will will still be there, namely the much more instant response time of LED's for brake lights and the realiablity of LED's especially in a bunch (if wired appropriately) makes the whole idea a safety benefit.

Mark

Filaments of bulbs used for automotive signaling actually have quite a fast response, and superiority of LEDs is mostly a myth. Paul Mathews

Na, I wont have that!! LED's shit all over fragile, energy wasting, slow responsive, more difficult to make and more expensive to make (if not already, very bloody soon) incandescent lamps!....... especially where brake lights and indicaters are concerned on a motorbike.

Mark

According to Hewlett Packard (now Avago), the advantage is about

120msec, i.e., 1/8th of a second. At a closing velocity of 20kph, that translates to 0.7m, hardly the dramatic difference claimed by the LED manufacturers. My own measurements showed more like a 70msec advantage for LEDs, but I'm not an LED manufacturer, just a photonics engineer. I'm all in favor of LEDs for safety lighting, but claims of dramatically improved reaction times are mostly marketing hype. Paul Mathews

I think what realy matters is a more effective warning signal, that the brain acts more quickly to. LED's lamps are startling. Avago who ? I have to go.

greg

This can't be true. You only have to drive around to see that LED tail-lights illuminate with a very distinguishable attack rate, much faster than incandescent lamps. In fact, I find them quite annoying because they trigger some kind of primal sense of urgency in my brain. I wish they would put a cap on the line, to slow them down, so they look light "normal" lights.

Perhaps not everyone can see this. I do have a high sensitivity to flickering monitors and PWM'd LEDs.

"Paul Mathews"

So you are talking about the human response and not the LED's response? That's a completely different matter then. Exactly what are you measuring anyway? The time from power application to full brightness, the time from power application to noticeable brightness? ....significant brightness? Exactly what are you trying to say here?

20kph is a very slow speed too. Most traffic around here drives in the 120kph or greater range. Considering that I've seen rear-end collisions avoided by mere inches on more than one occasion, I'd say that the extra 4.2m could make a difference. And at least for the time being (kinda like the third brake light in 1986) they _are_ more noticeable.

The average taillamp (1157) takes approx 220mS to reach 90% brightness. That's about 22' at 75mph (a much more realistic number). After all, that is more than a whole car length of extra warning time. I find that to be more than insignificant.

That means that the forth guy back might actually see illumination just as he slams into the car ahead of him. ;-)

While it's true that cars are traveling at 120kph and greater velocities, it's the 'closing velocity' that matters when you're considering the benefits of reduced reaction time. Closing velocities in traffic situations are usually more on the order of 20kph. I'm not saying there's no benefit to LEDs, just that the usual claims like a

4.2m advantage do not obtain in most traffic situations. The 'response' time claims are also full of hype. For example, you'll frequently see claims that LEDs are 'thousands of times faster'. This refers to the fact that LED response times are a microsecond or so, while filaments are in milliseconds. Using the ratio is pure hype, since it's the difference that matters. Even when considering the difference, it's worth noting that the 0 to 90% risetimes 'measured' by HP are not relevant to human reaction time, since, depending on conditions, humans can see brake lights even when they have only reached a small fraction of their ultimate brightness. When you actually measure reaction times (as I have done), you typically see about 100 ms advantage for LEDs. Calculate that advantage any way you like. Paul Mathews