When do you think we'll see 400-500 lumens from 2AA?

Hi,

Been away from the dedicated flashlight for a while now, with only one eye on it (last purchases were a SS LD01 and a black iTP A3).

I was wondering when the next big breakthrough might come (I know, I know - no-one really knows - but I'm an Apple guy and we thrive on rumour and speculation ).

I'm wishing that we see a 400-500 lumen 2AA light by the end of the year (aren't we all?).

How far off do you think my wish is? Another year or so? Maybe 2?

All the best,

Face

This is in principle possible, but runtime will be too little, and will need to ensure good heat.

Face said:

I'm wishing that we see a 400-500 lumen 2AA light by the end of the year (aren't we all?).
How far off do you think my wish is? Another year or so? Maybe 2?

Click to expand...

I guess two years.

I'd like a huge output 2AA light also, but I think the batteries are already being driven as hard as practical; some 2AA lights on max will empty a pair of Eneloops in about an hour so that's roughly pulling 2 amps, which is quite a bit. And that's only delivering about 700(?) miliamps to the emitter itself, so the boost circuit is eating the rest of the power.

A breakthrough would have to come from AA batteries that can supply a lot more power, a circuit that's much more efficient (although I think they're already 80+ percent), or my personal fantasy, an led that only has a forward voltage of something like 2 volts instead of the usual 3.6-ish.

All that being said, the current 2AA offerings are already pretty dang powerful with around 200 lumens OTF, which is plenty for most circumstances (remember when 2xCR123a lights wouldn't output that much only a few short years ago?!)

If I had to start from scratch I'd have a couple of 2AA and 1AA lights only, except for my only Li-ion light being an MCE blaster (but not for edc).

If you go with at the emitter lumens at turn on it's possible now.

The XP-G driven at 1.5A in an R5 flux bin can give 493 lumens. With a Vf of around 3.55 at that current you are talking greater than 2.8A draw though so don't try it on alkaline. Heat also becomes an issue in maintaining lumens Optical system losses pull you back under 400 OTF though. Still pretty amazing what's possible now.

Yeah, it's entirely possible, but it's also very impractical. As a result, manufacturers probably won't be rushing to make anything like that. To deal with heat (from both the converter and the emitter) the light would need to be on the bulky side for a 2AA light anyways.

Is the interest in seeing a light in that form factor or with a specific battery size/chemistry? If you are open to li-ions (14500, the AA equivalent) you could probably put together a 2x14500 light with an MC-E.

+1 for 14500s, 2xAA format would yield more like 600-700 lumens with the right LED, you'd just have to be comfortable with a Mag-size head on a MiniMag-size battery tube..

Hack On Wheels said:

Yeah, it's entirely possible, but it's also very impractical. As a result, manufacturers probably won't be rushing to make anything like that. To deal with heat (from both the converter and the emitter) the light would need to be on the bulky side for a 2AA light anyways.

Is the interest in seeing a light in that form factor or with a specific battery size/chemistry? If you are open to li-ions (14500, the AA equivalent) you could probably put together a 2x14500 light with an MC-E.

Click to expand...

Why are LED's so much more sensitive to heat than incandescents? Perhaps the answer isn't to build brighter LED's but rather LED's that can better tolerate heat?

Wait until Cree come out Cree XP-G T5.

That would be nice and let's see until the end of this year coz as this moment the market has a 300 lumen offering for 2xAA light. They can easily up the lumen up to 500 if they want to but that equals to very short runtime. Ofcourse batteries should be improve too inconjunction with this powerful lights.

As stated in theory , we should be able to get it now !

Whats needed is more powerful boost drivers , ATM were stuck @ 1A or so , and whats needed is 1.5A , 2A , and beyond .

old4570 said:

As stated in theory , we should be able to get it now !

Whats needed is more powerful boost drivers , ATM were stuck @ 1A or so , and whats needed is 1.5A , 2A , and beyond .

Click to expand...

Unfortunately, the more powerful boost drivers will mean that heatsinking the drivers as well as the emitters is an increasing concern. Improvements in efficiency are what I'll be hoping for as everything else will have some considerable compromises. (Though efficiency improvements mean less waste heat and then things can be pushed even harder... )

My solution to this was to use 2 14500s to run an MC-E in 2s2p. It's seeing 1A (approx 500mA per die)... but I still can't run it for very long without a considerable decrease in output. It's a fun toy though!
[Generally higher voltage and correspondingly lower current is more efficient... it also works better with the currently available boost driver technology, imo.]

Or we could hope for LEDs that only produce light in the visible spectrum (while we are at it in the same profile as sunlight) but without any electromagnetic radiation outside of the visible spectrum (esp. no infrared radiation). Shouldn't that be enough of a boost to efficiency to allow 500 lumens of perfect light from 2AAs?

And we would get rid of a big part of the heat issue.

I presume a large part of the heat comes from the die, and is perhaps generated by resistive heating. Whatever the cause, it means energy is going into heat rather than light. So an LED with a Vf that matches the batteries, and which does not generate (much) heat will greatly improve efficiency. I have no idea how efficient current LEDs are, in terms of the percentage of the input battery energy that is converted into visible light. But knowing that would give us the theoretical maximum output in lumens. According to Wikipedia, so it must be true, LED luminous efficiency is less than ~22%. Using a Quark 2AA as baseline, and knowing it outputs 200 OTF lumens for about 1 hour, tells is that a 2AA battery LED torch could output 500 lumens for one hour, or even 800 or more, if a suitable LED (or alternative light emitter) could be made. But in the real world, who knows.

I think it's about 250 lumens/watt for blackbody emission curves clipped to the edges of the human visual range. It's much higher for blue/yellow "phosphor white"

Chrontius said:

I think it's about 250 lumens/watt for blackbody emission curves clipped to the edges of the human visual range. It's much higher for blue/yellow "phosphor white"

Click to expand...

Surely an LED isn't a blackbody radiator. Hence it need not have lots of nonvisible radiation and can in fact have a very narrow emission spectrum.

recDNA said:

Why are LED's so much more sensitive to heat than incandescents?

Click to expand...

Well the whole method incandescents use to produce light is by heating the filament. Reflecting IR back to the filament to produce extra heating with less power input is a method of making incandescents more efficient.

If you do not mind missing part of the spectrum, quantum dot technology can double output of current LEDs. These LEDs just have blue from the die plus green and red quantum dots.
What about the other colours? Well both TVs and computer monitors only use red, green and blue and get away with it.