This is another important piece to keep in mind… I didn’t have any brand-new batteries to test. My alkaline was partially used.
Here’s Skilhunt’s chart. It probably pretty close to reality.
And yes… there are inefficiencies everywhere. Increased power draw decreases battery efficiency. The driver has some inefficiency (roughly 85% or so). You’ll lose some through the optic (~15% loss). It all adds up unfortunately.
The constant power discharge graph already accounts for that. Right? Falling voltage, rising current - it’s already there. When all is said and done Duracell promises you 750 mW for 0.5 hour = 375 mWh. Or, if you can lower you consumption rate, 500 mW for 1.1 hour = 550 mWh
It doesn’t look like like it’s 85. The LED consumes 221 mA at 3 V = 663 mW while in regulation. Once the light output drops I assume the power used drops proportionately, so I completely unscientifically integrate total power (663 mW * 5 min + 330 mW * 20 min) = 165 mWh. If we assume driver efficiency at 85 it’s 195 mWh out of the battery. So either your battery was half discharged (but then it wouldn’t measure 1.55 V) or the driver efficiency is about 50%.
I know even less about optics than I do about electricity, if that’s possible 
Let’s ignore that for now.
Thanks Gchart! Awesome data!
I did some crude and quick tests of my own E3A tonight with stock resistor HiCRI version with a bench supply and my eyeballs… I’ll likely do it again at a later time with more equipment and effort…
From off (sweep up), turn on voltage on my sample was 920-930mV. Once on, LED didn’t go dark until 0.13V!! I didn’t expect that!
From 1.4V down, I saw min input power at 800mW which increased linearly until about 1.25V ( Peak input wattage happened at 1.03Vin (1160mWin) after which power began dropping quickly. Heres the raw data in Vin mWin. Laptop is dying… edit later
1.4 0.84
1.35 0.85
1.3 0.87
1.25 0.9
1.2 0.92
1.15 0.97
1.1 1.02
1.07 1.06
1.04 1.1
1.02 1.16
1 1.09
0.97 0.99
0.94 0.89
0.91 0.79
0.85 0.59
0.8 0.46
0.75 0.34
0.7 0.23
0.65 0.14
0.6 0.07
Or, to put it visually
Thanks for that, I’m very much a visual person. So in the primary operating range, it appears to range between 800-900 mW. Certainly explains why alkalines really aren’t a great fit and that NiMH & lithium do much better.
If I’m honest with myself, idk if I’ll get back around to more measurements with this light this week. But I’ll clean up the post and data from my PC tonight. Thanks for posting that plot in the meantime Don.
Gchart, I agree that at stock output, alkaline is only compatible technically. Though in the 2.3 lumen mode and now seeing how far down this boost driver operates, I can imagine some crazy runtimes… And leak prone cells
Is that to be expected from a $10 flashlight? Or there are no boost drivers with >90% efficiency in 0.8-1.8 Vin range regardless of how much the light costs?
Thinking way back, my Walkman was able to suck two AAs dry in the late 80s, so it should be possible. Maybe not in a flashlight form-factor?
It’s all about the power draw. That’s just a lot of power to ask of a AAA alkaline. If the output were lowered a bit (like in my 42 lumen test), I imagine the alkaline would respond much better.
It’s hardly cell’s fault that the driver draws 1160 mW just to pass 663 mW to the LED, is it? It was doing it to the bench supply.
Took some comparison shots with some of my AAA flashlights. Not enough to open a new thread, hope that’s OK for you.
oooh, i actually been eyeing the edc01, cant believe its significantly bigger, hm…
edc01 (new worm), is a 3 mode light, arguably more versatile, and batteries can last longer
The E3a comes on at a brightness, 90 lumens, that drains an AAA in 30 minutes… to me, that is a waste… but its the smallest, lightest, highest CRI aaa option, if someone wants a keyring light for example. I personally choose not enlarge my keyring. My lights are easier for my use, separate from my keys.
the edc01 is rated to go 4 hours on its medium of 30 lumens, plus it has a 5 lumen low, plus a high that matches the E3a… however, there is NO High CRI edc01 option.
to get High CRI aaa, another option is the aaa Tool from drop.com… comes in fancy metal too… $lippery
so, for simplicity, the tiny E3a, with high CRI, is king of the little light Hill
try them all, then decide :money_mouth_face:
But it runs 60 minutes on an AAA. 
my mistake, you are actually correct… E3a are great little lights,
zeroair did a runtime and review
and it did last 60 minutes on an AAA, which is impressive :+1:
Runtime on these is impressive! I’m flummoxed that the new Skilhunt E2A AA can’t keep up despite more capacity and larger host. Zeroair’s review shows the E2A is stabilized at only about 65 lumens.
Thanks gchart for these very interesting runtime graphs.
I wonder how a primary lithium would behave: if any of the "super-reviewers" has one at hand, I'd like to see it too on a graph.
I have bought a couple of E3A as gifts (and will buy more when I find a super deal), and plan to include a primary lithium in the gift, for people who don't have rechargeables and a charger.
OTOH, here are the result of a few attempts to improve the beam, since I didn't like the beam irregularities from the stock pebbled optic (though only visible in "white-wall hunting"). I've tried the following lenses from Yajiamei Store:
Smooth lenses:
Frosted lenses (they call them "Shamian"):
In conclusion, if you like smooth beams without significant color shift, do your E3A a favor by replacing its pebbled lens with a 20-30° Shamian one, or 60° if you like a more "wall of light" beam.
Unfortunately I was not able to take nice beamshots like some of you do, my attempts did not do justice to the beam improvement, but my eyes do see the difference.
Cemoi, thanks for the write up. Useful information for sure.
Personally, I find the stock beam nearly flawless and very pleasing. May I ask you what issues you saw?
See this post in another thread, and posts #55 thru 58.
But as said earlier it is only visible at short range (1' from a white wall), by picky flashaholics like me 
Different LEDs I guess.
E3A is using the Samsung, E2A the SST20.
Emitter efficiency is probably part of it. The E3A he tested was neutral white, but not high CRI. The E2A is 95 CRI. That should cost it around 30% efficiency.
It’s more than that, though. Zeroair’s graph for the E2A shows about 50 lumens/W on medium and 63 lumens/W on high.
His E3A graph works out to around 120 lumens/W.
Even considering the SST-20’s are not the most efficient emitter around, I would have expected them to be able to achieve over 80 lumens per W on high with a comparable driver. And a stepdown should not be needed at all on the E2A when running on AA batteries. It seems to use the same logic as with 14500 batteries.
I have an E2A and am happy with it - the size and hand feel are great, and the optic they use makes a superb beam profile in my opinion. It’s not the light to get if you need maximum output or runtime, though. It’s a good light that would be great if they improve the driver.