Why there is such a huge difference in runtime? 1 AAA VS 2 AAA

I was comparing two identic lights but different battery configuration.

http://www.lumintop.com/iyp07.html only 4 hours at 40 lumens
http://www.lumintop.com/iyp365ti.html 25 hours at 40 lumens

Why it is almost 5 times less?

Only thing I can think of is that because both lights are boost converters, they have to suck in more current than’s supplied to the LED. With 1 AAA, that’s at least twice as much current and a hard boost, vs 2 AAAs and a mild boost.

Alkaleaks have high internal resistance (which is why those crappy 3×AAA lights at a nominal 4.5V won’t fry a handful of dinky 5mm LEDs at ~30mA a pop with fresh cells), so more current sucked from a single cell means more voltage-sag, and lower terminal voltage, and cutting off lots earlier than if 2 cells were used.

I don’t believe those numbers. 40 lumens for 25 hours is 1000 lumen-hours. They don’t say what batteries they’re using, but if they’re Ikea Ladda 900s with 1.158 Wh each, for example, that’s 432 lm/W. That’s more than double the efficiency of the best I’ve seen from any flashlight, and pushing, if not beyond theoretical limits for the efficiency of white light sources.

If it runs for 25 hours, it’s not staying at 40 lumens the whole time.

Yeh, it’s probably “ansi lumens”, ie, time it ’til the output drops to 10% the initial value.

But that’s the point, with 2 cells, sucking down half the current is more sustainable, even at the tail end.

I’m just running a few hypothetical numbers here. This is a xp-g3 test done by TA.

Given the 1AAA is making 40 lumens for 4 hrs.
That should be about 100ma at the led but because of the boost driver voltage conversion thats about 200ma from the battery, increasing as the voltage drops.
200ma for 1 hour is 200mah’s. So 4 hours needs about 800mah of capacity.
That seems about right when you factor in the losses in the driver from the boost conversion.
.
Using 2 AAA’s you would still need a boost driver given the voltage will be below the vf during the discharge but not nearly as long.
So you wouldnt have as much battery energy lost during the run time because of the higher voltage to less current needed for the conversion.
But both running at 40 lumens the led has to be driven at the exact same current. So basically the only difference is you have doubled capacity with better efficiency.
So I’m just guessing with 2 AAA’s the light should run about 9 hours at 40 lumens.
Just my 2 cents :slight_smile: .
.
For 25 hours at 40 lumens, we know the led needs about 100ma. If the input voltage was dead on vf of the led and didn’t drop not even needing a driver.
1 hour at 100ma = 100mah
25 x 100mah = 2500mah
The 2AAA’s run in series has now become a single battery of 3v output with the same capacity as one AAA, about 900mah.

No, it’s not. I measured 560 minutes (9.5 hours) on medium on my IYP25.

Side note, I should really update my plotting script to display runtime in hours:minutes instead of just minutes.

It could be anything.
Most likely the 40 hour one is just wrong.
Or intentionally misleading.

I’ve never done a runtime test on my IYP365, but it sure seems to run a long time on pair of Energizer Ultimate lithium primaries. I prefer them because of their perceptibly lighter weight.

I use it frequently, for brief intervals, every day and it’s still running on the same pair of cells I loaded in November. Although I end up using it almost exclusively on ‘medium’, I just now clicked it through its modes and it still offers a good strong ‘high’.

I find that a 2xAAA light in my shirt pocket is indispensable, as 90% of my applications are for a minute or two of modest illumination indoors. I have plenty of ‘barn burners’ for the outdoor / night time applications, but this instantly accessible IYP365 is undoubtedly my most frequently used light. I spend much more time looking for that tiny screw I dropped under the work bench than I do looking for coyotes out in the pasture.

With my AA and AAA lights, I simply change cells when ‘high’ is about the same brightness as ‘medium’. That generally keeps me out of an unexpected ‘uh oh!’ and sudden darkness. Hillbilly voltage testing, I suppose, but it works for me.

Yeh, that’s how my Quark worked. When high = medium, that’s my early-warning. When high = medium = low, it’s time to change the alkaleaks unless I wanted to ruin the light.

Yes its an amazing light for casual use, the batteries will last forever for just short use.

Great. The shorter version will be then far more than enough for casual use.

runtime calculation shenanigans aside a 2aaa carries 2x the energy.
and the current is lower which also works in its favor. Peukert's law - Wikipedia