longest running AAA light?

I am intrigued by super long run time lights, probably because I am a bit of a cheap skate. I was reading another "thread":https://budgetlightforum.com/t/-/35362 recently that discussed getting a week or run time on a Manker Boney modified by changing a inductor. Having a electrical engineering background, I would love to understand how the inductor value change produced this effect. I have some ideas, but I digress . .

Anyway prior to reading the thread above I happened across a thread on the CPF regarding longest running AAA lights, so I decided to start a simple test of my own. I have two AAA (manker e01 and fenix ld01) lights which I am testing, along with my petzl e+lite (two cr2032 button cells). The E01 advertised 100 hours on low, and the e+lite advertises 75 hours on low. We will see who wins. The first run is on eneloops in the e01 and ld01, which I don't expect to last as long as the manufacturers claims. Then will come the alkaline tests. Of course the e+lite takes lithium coins, so I will only do one run here. I am comparing the output to my thrunite archer v3 as a reference, which boasts 17 days on low, but runs on an AA cell. Would love to hear about other's run test on AAA lights as well.

Summary

LD01 on eneloop 32 hrs still at full output, 36 hrs it was off. Not sure when in between it went out.

Fenix LD01 on alkaline, a few year old one starting at 1.53V 8hrs 1.37V 21hrs 1.28V 33hrs 1.17V. Lasted 38.5 hours on low, with 4 times turning it off to measure the battery, then back on to medium, than switch quickly to low again. Not bad compared to 27 hours advertised by Fenix.

Manker E01 on eneloop 56hrs 1.28V 78 hrs 1.27V 91 hrs 1.27V 103 hrs 1.27V 116 hrs, 1.26V 142 hrs 1.25V 166 hours 1.23V - thinking this is getting near the end . . . 190 hours 1.21V 200 hours 1.16V, test over. light was still usable in the middle of the night. I don't want to keep discharging the eneloop, and also want to start investigating the current vs voltage profile. I put a new alkaline in it today and found that it is as bright as the AA Archer which also claims 1 lumen on moonlight. The eneloop was never that bright, so it seems that the output is not regulated on moonlight is the E01.

Petzl e+lite just illumination qualitative estimates 24 - 56 hours dimmer than new, but much brighter than EO1 spot on wall. e+lite is much more a flood light with no reflector between 56 and 70 hours dropped off to just brighter than E01, still that bright at 78hrs. 91 hrs now dimmer than E1, but still can read by it. 103 hrs, pretty much the same. e+lite still barely producing light at 142 hours, not terribly usable, but light none the less.

Thorefire PF03 with 750mAh eneloop, between 26 and 27 hrs

Lumintop tool The aluminum XP-G2 version has expired between 23-26hrs

Lumintop tool copper Nichia 219b lasted between 28-29.5 hrs until dead

Lumintop brass worm 31 hours on eneloop. A bit surprised it was shorter than the LD01, and the advertised time is longer. Worm lasted 34.5 hours on 4 year old alkaline battery, same vintage and starting voltage as used on the Fenix LD01 test above. Fenix wins.

I’m also interested in this as I just recently stocked up on some good clicky AAA lights for EDC. I plan to do runtime tests on:
Lumintop Tool (Al XP-G2 & Cu Nichia 219b versions)
Lumintop IYP365 with 1xAAA (head will fit Tool body)
Thorfire PF03
Olight i3 EOS (older light for reference)

I expect the Thorfire to last the longest in low, it is truly a firefly mode and you can look directly at the LED without issue. The IYP with 1xAAA looks like it lowers the med and low modes, while the high appears close to the other Tool, but I am only estimating visually.

The Thorfire claims 240 hours on low!!! If you confirm anything close to that, then that would even longer than the Manker Boney in the thread I referenced above. I wonder if they “accidently” added a zero to the end. Let us know!

btw, the LD01 is 23 hours in on a white eneloop AAA and it looks good as new best I can tell.

The Reylight version of the copper Tool, which has a lower low (I measured 1.6 lumen) than the normal Lumintop Tool, ran for 100 hours to 10% output (this is on a Duracell alkaline):

It achieves this thanks to the low output, the driver is not one of the most efficient ones.

Based on the little I know about lights it seems like the reylight copper tool puts out a decent amount of light for 90 hours. definitely more than my E01 is putting out.

47 hr update. I am figuring this out as I go, and I don't have any fancy equipment. I am using my cell phone camera and a DVM. By all data I have the LD01 is still running the same brightness as when I first took data, which is about 7 hours in to that light being on. According to "this":http://lygte-info.dk/review/Review%20Fenix%20LD01%20SS%202010%20UK.html the output is regulated down to about 0.5V. I took the battery out today after ~29 hrs and it is at 1.21V. The LD01 has now been running for 32 hrs and looks the same in pictures as it did at 7 hours. I am thinking/hoping it will go all night without killing the eneloop, time will tell. Nope, it was out cold at 1:40 am. Verified 32 hrs and a few minutes. Not sure when it died. Eneloop measured 0.83V and slowly rising when I checked at 1:40 am. Put it right on the charger at that point.

The e+lite has dimmed from when I started the test, but not much if at all from the 24 hour mark. It puts out more than enough light to find my way around my darkened bedroom without dark adjusted eyes. If I used it as a head lamp it is easy to read by. 56 hours so far on the e+lite and it looks iike it did at 24 hours, still fairly bright.

And the Manker E01 is very steady, even though somehow it got into medium mode for ~10 minutes yesterday. A few hours ago the battery was at 1.29V. This will go for a couple more days I suspect easy. Good thing, as I am leaving on a business trip in the morning. I likely won't bring my DVM, but I will bring the lights in the on state packed away in my bag somewhere. I will do a check of the Manker again before I pack up. Measuring the e+lite is a bit of a pita, so I will just monitor with my cell phone camera. At 56 hours no change in brightness of the Manker E01, or should I say dimness, at 0.1 lumen. Battery at 1.28V. This 0.1 lumen is not as bright at my thrunite archer 0.1 lumen, and was not on a new battery either. No change in 56 hrs.

I’d be more interested in max lumen-hours.

Any light can last forever if the output is low enough.

Also I’d like the light output not to vary over the lowering battery, until it has nothing left.

I can see that. I can only go by the advertised lumens of the LD01, and it ran for over 32 hours at 3 lumens. Since I am not sure when it started to fade this is just an estimate, but 100 lumen-hours with a constant output. Not too bad for a AAA.

Thrunite AAA flashlights are know for very long runtimes.
ThruNite Ti XP-L claims 115 hours with 0,5lm output.
Older models (Ti3) have even lower firefly mode around 0,04lm but also older LEDs so the stated runtime is 115h as well. With good alkaline, or better yet with lithium AAA cell, this runtime should extend further.

Disappointing result from the Thorefire PF03 with 750mAh eneloop, between 26 and 27 hrs before it was dead. I was hoping for at least a week, maybe it would run longer on alkaline batteries.
Now testing the Tools with the same batteries.

I thought 240hrs was too good to be true at that level. Thanks for sharing the data.

Google search “joule thief”. Can light an LED while running the battery down to ~0.3V, well beyond when it is “dead” in any other light. Can run for weeks.

Nice! Thanks for the tip. I now understand how changing the inductor in the link referenced in the OP changes the output. I will be gathering up some parts to make one of these when I return home.

The main problem is that at low loads the boost converter is not efficient anymore,
bacause the power to maintain it is not much smaller than the power needed to drive the LED,
So the runtime between 0.5 and 0.05 lumens wont be anywhere near 10 times longer

I bought at Thrunite Archer V3 a couple months ago, probably should have considered the one you recommended also. At the time I was looking for a AA light, now I feel a AAA would suit my general EDC needs better. Or a AA that is much simpler (and smaller).

70 hours into the test and the e+lite is definitely fading. Since I was traveling all day it is hard to say where between 56 and 70 hours it dropped off. But now it is just a bit brighter than the Manker E01. I was using the e+lite to read a map while traveling tonight, so it is still useful. Same brightness 78 hrs in.

And somehow the Manker E01 battery is still at 1.28V. I checked the light a few times today to verify it stayed on, but somehow it is bright as ever and no change in battery voltage. Not sure how sensitive the voltage of a eneloop is to temperature, but it is a bit warmer here than where I measured it early today, probably 5-6 degrees C. 1.27V at 78 hrs.

restarting the LD01 on an alkaline battery now, 11pm EST. Fresh battery measured 1.53V. 1.37V at 8 hrs into run with alkaline.

The aluminum version has expired between 23-26hrs, but my copper version is still running at 27 hrs. I believe the copper to be the ‘efficient driver’ per this thread Efficiency Copper Lumintop Tool, Worm, Maratac, ReyLight
When running them, I’m recording the time to when there is no light from the LED without interruption. I know sometimes if a light is turned off and the battery rested then it can give a little more, not sure about the science here but I’m sure someone does.

Thanks for the data! I have not looked up the tool output levels, I am curious to hear how it compares to the 3 lumens advertised on some version of the LD01. Likely not my version, as I have been able to find out very little about this particular emitter, the XP-E R4.

As far as not shutting of the light goes, I thought about that but decided it is a bit more real world than just leaving a light on until it dies. Also, I don’t have a way to quantify performance other than camera beam pictures, which are a bit of a pain to load on this site.

91 hours into E01 and e+lite runs. They are about the same brightness now. The E01 battery is still at 1.27V. If I was not measuring this myself I would not believe it. The light intensity has not changed from the test start for the E01. The LD01 is at 21 hours, battery voltage is at 1.28V. No drop in light output yet for the LD01. I hope to be sleeping at the 100 hour mark for the E01 and e+lite, which will be the 31 hour mark for the LD01. I suspect by the time I get up they will still be running though.

So the LD01 lasted 38.5 hrs on a alkaline battery. I took the battery to the meetings I was in for work today and peeked at it periodically. I was able to see it fully illuminated and then 10 minutes later looking dark. When I returned to the hotel in the evening I turned it back on and it produced full level low light for about 40 more minutes. It slowly died down to barely discernable light, and then I removed the battery and quickly measured. The battery had recovered to about 0.38V and rising. The review link below shows behavior of a version of the LD01 that looks very similar to what I observed. Operation down to 0.3V. I am thinking I don’t need to make a joule thief, I have one already in the LD01, which seems very efficient. I can’t verify the intensity level of this light but if the advertised 3 lumens is correct, then I got about 115 lumen hours out of a 4 year old alkaline battery. And the light does come back on at medium level resting during the time it took me to type this post.

I also did another voltage test of the E01 battery. It is holding tough at 1.26V after 116 hours. Just as a sanity check I put a new eneloop in to compare the light intensity which best I could tell had not changed since the test beginning. I can see that there is some slight dimming of the already dim light, but not much. The light is still useful finding your way around a dark room, still puts out enough light to read by, but maybe not fine print. I also found out that the mode memory timeout in the light is a function of the battery voltage, which is no surprise. When I took out the new eneloop and put ack in the old one, I observed (twice) that the light came on in medium mode with the old battery, and I had to switch to high then back to low mode. The first time I thought the light was just flaky, but after the second time I realized it is the short term mode memory. I tried two more times, both times waiting well beyond 10 seconds, and it came on at low in those cases. Also, it never came on above low when I put the new battery in. Fun stuff!!

I leave for the west coast way too early tomorrow morning, so I may not check any voltages for a day or so. I have no doubt the E01 will still be on, we will see about the e+lite, which continues to very slowly dim but is still putting light out.