[Closed] FAO: Trout / People who build bike lights / LED stuff

Hi JT
the Insane mce has 16 dies as each mce has 4 dies so each mce kicks out 800 lumens x 4 giving the 3200 insane ness

Each single XP-G will kick out 345 lumens which if x 16 dies would be 5520 lumens.

so we now only need 3 dies to better the MCE / P7 s

I was all for the lumen race last winter but found that 1200 on the bars & 1000 on the lid to be plenty - only used on full power for downhills & reduced to less than half power for climbs to extend battery life.
Improving beam pattern & battery life/reducing size are my priorities now I've enough lumens to not limit my speed.

Trout, you're a gent!

thank you 😉 very much appreciated indeed 😉

5520 lumens, that's got to be like main-beam on your average car?

Snaps: my thoughts were to one or two LED units, but with these more efficient ones, the run-time can be increased, giving me a longer burn time = more time riding (i'm not the quickest rider!)

And, I have a vested interest in using these for some other none-bike related idea's, whether they come into fruition is another story though 😉

So in theory, I could use 2x of these XP-G LED's, on approx half power (350mA per die) giving, i think, a much more efficient use.... lower heat levels = more of the energy is converted in to light and not heat...

or have i got that wrong!?

thanks 😉
Jonathan

LED's are actualy more efficient the higher power you run them at. So a 100lumen LED, running at 50lumen might be using 60% of the power it would at 100lumen. So 1 LED at 100% is better than 2LED's at 50%.

The problem is that althought they'r 4ish times more efficient than a halogen bulb, that still means a lot of energy gets wasted as heat. This is why small LED's with big heat sinks are better in bike lights than the big LED's that you used in D&T at school.

so we now only need 3 dies to better the MCE / P7 s

But then the die area of the XP-G is twice that of an XR-E/MCE/P7, so it actually puts out less lumens per surface area than any of those. Meanwhile if you want to beat an MCE/P7 light you need more individual LEDs and optics, thus making the light bigger and more complex. Sure it's an advance in efficiency, but I'm still not convinced it's the big leap forward some seem to be making out (certainly not in the same way the original XR-E with it's instant doubling of efficiency in the same size die was).

LED's are actualy more efficient the higher power you run them at. So a 100lumen LED, running at 50lumen might be using 60% of the power it would at 100lumen. So 1 LED at 100% is better than 2LED's at 50%.

Completely and utterly wrong!

This is why small LED's with big heat sinks are better in bike lights than the big LED's that you used in D&T at school.

By big LEDs you mean a standard 5mm domed LED? It's a long time since I went to school, but we did have LEDs back then, and that's what we'd have used and what I would imagine they use now, given anything normal bigger than that is a specialist item. You do realise that 5mm LEDs actually have smaller dies than the power LEDs we're talking about?

no, i'm failry sure i'm right, LED's become more efficient the more current you pass through them. Just as normal diodes resistance drops the more current passes through/voltage is applied to them.

no, i'm failry sure i'm right, LED's become more efficient the more current you pass through them.

Based on what evidence? Have you even ever looked at an LED data sheet?

Just as normal diodes resistance drops the more current passes through/voltage is applied to them.

Given you don't normally describe diodes as having a resistance, I suspect you really don't have much idea at all. The fact that current in a diode increases exponentionally with the applied voltage is also totally irrelevant to the efficiency of LEDs.

LED's are actualy more efficient the higher power you run them at. So a 100lumen LED, running at 50lumen might be using 60% of the power it would at 100lumen. So 1 LED at 100% is better than 2LED's at 50%.

Not true - see the datasheet (the graph on page 8 is what you're looking for)

This is why small LED's with big heat sinks are better in bike lights than the big LED's that you used in D&T at school.

No it isn't, it's because the power leds are able to be run at 1 amp, whereas the LEDs you used at school are running at .02 amps (and will blow up at about .05 amp)

Joe

fight fight fight ... oh, not in school ....

I didn't know that about running them at a lower current being inefficient!

I haven't seen the spec sheet for the XP-G but would imagine that if its larger than its counterpart (the XP-E) then it would be of similar size to the P7 unit?

My thoughts were to use 3x XP-"X" (X being either E or G) to give me a sizeable amount of luminosity wherever I need/want it.

But it's the efficiency bit that interests me too, so it would seem 2x LED's at their max current is more efficient than 3x LED's at 50% current ...

interesting 😉

But it's the efficiency bit that interests me too, so it would seem 2x LED's at their max current is more efficient than 3x LED's at 50% current ...

Nooooooo! Don't listen to him - as I keep trying to point out, he's totally and utterly wrong.

light output is roughly proportional to current

current is exponential to voltage (i actualythought it was a dog leg with 2 straight ish lines?)

so once you get past the dog leg (or if its exponential it just gets better all the time) then power (current x voltage) is increacing much slower than light output (proportional to current) ?

happy to be proven wrong if soene can plot the power vs light output graph?

i stand correced (just been playing in excell)

efficiency does indeed drop with power.

Turns out my old a-level teacher was wrogn after all!

and why would you be bothered about "efficiency" in the sense of carrying round more lights to get the most out of a small battery, surley its easier to stick another battery onto one light?

Running the leds at lower current is more efficient! than running at higher current but at high current you will of course get more lumens and loads more heat and less runtime .

2 XP-g @ 500 ma will be better than 1 XRE @ 1000ma
and make a smaller light with a better tuned beam using 2 small optics
rather than 1 large optic.

so once you get past the dog leg (or if its exponential it just gets better all the time) then power (current x voltage) is increacing much slower than light output (proportional to current) ?

Glad to see you've worked out you were wrong, but I should point out your A-level teacher may not have been - you might juist have misunderstood him. You see, if light output were proportional to current, then [b]any[/b] increase in voltage with increasing current would mean it became less efficient the higher the current, since the power is actually increasing faster than the current, not slower as you assert.

Of course in the real world light output isn't proprtional to current in either, hence it gets even worse.

Oh, and many/most of us don't want to have to carry round the extra battery to stick in, hence why effiiceny is important.

thanks trout (and others) 😉

most appreciated 😉

the battery I have was built by the very capable battery builder, smudge! 3.7v @ 7500mAh - and i'd like to use this battery, meaning I can save costs a little.

Just got to sort out the other things like connectors, oh, and whether I can build a much higher capacity battery than the 7.5Ah unit I have now.... but it appears its unwise to use more than 4x 18650's as one unit.

thanks again,
jt

whats that battery built from?

my current setup is 4x Q5's emiting 240lumens each running off 2.2V 2900mAh battery packs (can be wired up to give 11600mAh at 2.2V). But I'm on the lookout for something lighter and less bulky and simpler to charge than the 8AA's its built from.

thisisnotaspoon: it's built from 3x 18650's ... it fits very well into an old cateye battery pack (from an abs35) its the long thin tubes designed to either be bar mounted or as a twin pack under the bottle cage.

wired in parallel of course 😉

smudge built it (from here) about 35quid all in including a charger and assembly, top quality job - very pleased with his workmanship and service indeed 😉

many thanks jontown, please email me with any bigger capacity batteries youre after (up to 10.4Ah) 8) leave the lights to troutie he knows best. Who else gets up on the yorks moors at 4-5am for beam shots! dedication I call it.