AA battery tests incl. freezer test

[Egsise]

Test method:

-Tested cells have been discharged/charged about 10 times, exceptions will be mentioned.

-Charger is Sanyo MQH02 with 1090mA current which has charge completeness of around 85% compared to 0.1C charger.

-Batteries will be charged appr. 6 hours before freezing, batteries are inside the flashlight during freezing.

-Flashlight with batteries is freezed for appr. 24 hours.

-Room temperature test is made appr. 32 hours after charging.

-Freezer temperature is -22ºC (±2ºC) / -7ºF (±3ºF), room temperature is +22ºC (±2ºC) / +71ºF (±3ºF)

-Light output is measured with homemade lightbox, luxmeter is DX model sku.5100

-During the runtime test the flashlight is in room temperature, flashlight is turned on 30sec after taken from freezer. First reading is after 30sec runtime, after this webcam takes picture of the luxmeter every 60sec.

-Whole test is made without touching the flashlight before it reaches ~2% brightness.

-Flashlight is cooled with a fan, in freezer test fan is started after 30min runtime.

-Runtime is counted from the first reading to 50%.

-The output readings are not lumens, just relative total output numbers.

-Test flashlight is Fenix TK20:
Serial Number: XXXXXXXXXXX
Model: TK20
LED: CREE Neutral White Q2
Reflector: Smooth Reflector
Color: Black body with yellow rubber grip
Version of Warranty Card: English
Sales Region: Europe Area
Sales Country: Finland

-TK20 amp draw is around 1.5A.

-Test equipment, rechargeables and batteries are mine, exceptions will be mentioned.

-Test results in post #2

[Egsise]

Results in random order.
Click for bigger cell pictures.
----------------------------------------------------------------------



************************************************** ******

Sony Cycle Energy Blue cells Made in China = RayOVac Hybrids.


************************************************** ******

Sony Cycle Energy Blue cells Made in Japan = Sanyo Eneloops.


************************************************** ******



************************************************** ******



************************************************** ******



************************************************** ******



************************************************** ******



************************************************** ******

Coming: 5y old NiMH etc.

[Egsise]

reserved

[VidPro]

quick question, how long during the discharge are they actually at the -22C? beings they will make thier own heat at discharge.

i am not suggeting that a lightmeter would work in the freezer.

[bretti_kivi]

I think a -20 test would be possible in winter, but I suspect that the lightmeter really won't work at that temperature.

Bret

[Egsise]

Originally Posted by VidPro

quick question, how long during the discharge are they actually at the -22C? beings they will make thier own heat at discharge.

i am not suggeting that a lightmeter would work in the freezer.

The point of the freezer test is to see how low the output is before the batterys internal resistance, and flashlights driver circuit heats up the batteries.

So the answer is: not a second, all measurements are done in room temperature.
Measuring starts 60sec after the flashlight is taken from freezer, and the whole test is made without touching the flashlight before it reaches ~10% brightness.

The batteries start to heat even if the flashlight was in freezer the whole time, so there was no point making the testing very difficult.
And, in real life situation the flashlight is propably in the users hand.

[VidPro]

Originally Posted by Egsise

The batteries start to heat even if the flashlight was in freezer the whole time, so there was no point making the testing very difficult.
And, in real life situation the flashlight is propably in the users hand.

i was just trying to match up the cold numbers with the data graph info, because later they start doing just fine.

but if i was Stuck out in -7*F temps , and i have been in temps worse than that in whyoming out working. i wouldnt be holding a -7*F flashlight with my hand :-) unless i wanted it stuck there and going to the hospital to have it and my skin removed humans would have nice fluffy gloves on in temps like that. and with the chill factors we had, or winds buffetting everything, nothing warmed up.

then i wouldnt take a flashlight that only ran for 60 minutes , and if i did i would be running it in shorter bursts, or at a lower level.
sooo
i was just saying that if the test tries to show what its like out there when your freezing your backside off, and need light too, then the first 10 minutes Really show some of the best stuff.
Although all of it is good stuff too.

[Egsise]

Originally Posted by VidPro

i was just trying to match up the cold numbers with the data graph info, because later they start doing just fine.

but if i was Stuck out in -7*F temps , and i have been in temps worse than that in whyoming out working. i wouldnt be holding a -7*F flashlight with my hand :-) unless i wanted it stuck there and going to the hospital to have it and my skin removed humans would have nice fluffy gloves on in temps like that. and with the chill factors we had, or winds buffetting everything, nothing warmed up.

then i wouldnt take a flashlight that only ran for 60 minutes , and if i did i would be running it in shorter bursts, or at a lower level.
sooo
i was just saying that if the test tries to show what its like out there when your freezing your backside off, and need light too, then the first 10 minutes Really show some of the best stuff.
Although all of it is good stuff too.

Of course your hand would be covered with something, in -22ºC it is normal that you wear clothes, including gloves etc. in your hands.

Chill factor, no matter how strong the wind would be, it would still be just -22ºC to flashlight.
And your flashlight would be covered by a fluffy glove, so no wind chill...

I agree with you that my freezer test is good only for the first few minutes, but since I didn't find any similar tests this is better than nothing.

I started my tests because vee73 has tested AAA batteries with low drain 2xAAA direct driven flashlight.
He included a few LSD NiMH cells in it too and the results were scary, flashlight with Eneloops didn't even glow after it was frozen for 9 days in -19ºC.
His test results can be found in here.

Because I live in northern Finland, -22ºC is what I would describe " a normal winter day", so I expect my flashlights to work in that temperature.
Without lithium batteries of course.

[VidPro]

Originally Posted by Egsise

I agree with you that my freezer test is good only for the first few minutes, but since I didn't find any similar tests this is better than nothing.
.

it is excellent at any rate, cant wait for the rest of the results.

[Gordo]

Too bad that there isn't a test where you could leave the flashlight in a cold environment with a 10-min on followed by 20 min off cycle. Start at room temp (baseline). then after 20-min at test temp. Repeat until output reaches 50% of baseline.

Anyone got a remote control for a Fenix?

[ltiu]

Nevertheless, the test is relative and it shows that Eneloops are the best.

[Egsise]

Originally Posted by ltiu

Nevertheless, the test is relative and it shows that Eneloops are the best.

So far.

[Egsise]

Originally Posted by ltiu

Nevertheless, the test is relative and it shows that Eneloops are the best.

Oh and I hope you looked at the AAA test...

I started my tests because vee73 has tested AAA batteries with low drain 2xAAA direct driven flashlight.
He included a few LSD NiMH cells in it too and the results were scary, flashlight with Eneloops didn't even glow after it was frozen for 9 days in -19ºC.
His test results can be found in here.

[ltiu]

Originally Posted by Egsise

Oh and I hope you looked at the AAA test...

Cool! Thanks for sharing.

[Egsise]

Updated post #2, added Energizer Lithium results.
My TK20 must have really high Vf led in it, the runtime with lithium was ~25% less compared to other runtime tests what I have found.

Strange, very strange.

Anyway, with frozen lithiums the TK20 dropped out of regulation after 20sec, regulation jumped back on 20sec later so thats why the starting brightness is so low.

I also found Sony Cycle Energy Blue cells that were made in Japan, so I'll test those too.
The shop where I found them had both Sony cells, Chinese and Japanese Sony Cycle Energy Blue cells were happily in the same shelf.

[ltiu]

Originally Posted by Egsise

Updated post #2, added Energizer Lithium results.
My TK20 must have really high Vf led in it, the runtime with lithium was ~25% less compared to other runtime tests what I have found.

I don't understand, compared to what tests you have found???

[Egsise]

Originally Posted by ltiu

I don't understand, compared to what tests you have found???

My TK20 in roomtemp with lithiums: 1:34 to 50%

Tk20 runtime with energizer lithium batteries on High (about 2:05 to 50%)

Tk20 runtime with energizer lithium batteries on High (about 2:11 to 50%)

[Bones]

Many thanks for conducting these tests Egsise.

Considering how prevalent they've become in our stocks, it's good to see the Eneloop proving itself once again.

Thanks as well for the images of the two varieties of the Sony Cycle Energy Blue cells, I didn't realize until now just how closely they (externally, at least) resembled each other:

-

[firefly101]

Egsise, thanks for your tests I DO hope you do more for cold weather. I too live and work in a very cold area with Temps similar to yours.

I am REALLY impressed with Eneloop batteries performance, To so that Stable line with no dips is impressive, especially when Lithiums are usually the BEST for cold. I guess the changes they made to the internals are closer to the lithium batteries and the chemicals in this case are the NOT the important factor.

Are you going to do multiple tests to get an average?

Pretty soon you will be able to stick the flashlight outside to see what will happen. lol

[Egsise]

I have no intentions to do these test more than once, I have no time for that.
And I dont want to buy several sets of different cells.

Why?
Because AA cells are kinda pricy in here...

Cheapest online store prices in Finland 10/2009 for 4cell pack:
GP ReCyko+ 25.00$
Sanyo Eneloop 20.69$
Energizer Lithium 11.69$
Duracell Plus 7.25$

Fenix TK20 costs 85.00$

1 litre of gasoline 2.00$

1 litre of milk 1.40$

Average income is 40 000$ per year.

[firefly101]

Well except for the batteries your doing better than here.

$3.50 milk
$2.87 Gal Gas

Average Income in US $25-$35,000

Plus our dollar is about to be dump as the reserve currency so we should really enjoy the high prices coming. lol

I understand though, its just 3 tests should be done to make sure you dont have a Fluke in your results, example being you may get better or worse results on a second run through.

[Egsise]

Umm, 1 litre is about 4 gallons.


Yes I know that these tests should be done with multiple samples and several times.
But on the other hand, the test I did is so easy to do that anyone can do it.
All you need is 2xAA flashlight and a freezer to get rough results, you don't need a lux meter or lightbox to notice if the output drops from 150 to 50 lumens.

It would be great to see similar tests done with AAA, CR123 and 18650 Li-ion cells.
My brother has a Fenix TK11 R2 and I am planning to freeze it when I find the time for it.

[firefly101]

I was just showing you cost of living is more here in the USA, but then we get all the otehr stuff cheaper. lol Mainly because we have no import taxes like Europe does. You get more gas for your dollar than we do.

As for the tests I only have Duraloops available to me besides lithiums and no super flashlight to test in, I havent bought one yet, though I plan to later.

[MichaelW]

Any chance for a general mode test, maybe not a full duration but one hour to see thermal effects.

[TONY M]

Thanks Egsise for this useful info, this is great. I really am surprised by how well the "loops" did at low temperatures.

Originally Posted by firefly101

I was just showing you cost of living is more here in the USA, but then we get all the otehr stuff cheaper. lol Mainly because we have no import taxes like Europe does. You get more gas for your dollar than we do.

I don't think we get more gas for dollar than you do in the US thats for sure lol. About $7 per gallon right now. Its just not fair.

[Egsise]

Originally Posted by firefly101

You get more gas for your dollar than we do.

Nope.
US = 2.87$ for gallon of gas
FIN = 7.58$ for gallon of gas

Originally Posted by MichaelW

Any chance for a general mode test, maybe not a full duration but one hour to see thermal effects.

Yeah good idea, i have to do a few tests with alkalines, if they drop out of regulation then I'll do the tests with NiMH also.

[txmatt]

...from different thread but referencing this one...

Originally Posted by Egsise

Like I said, read the whole thread to understand what was tested and how.
I did not test runtime.
The tests were made in room temperature....think about it.

And trust me, you definitely notice when your flashlight output drops from 150 lumens to 50 lumens.

Could we continue this in some other thread, this is so off topic.

It IS a runtime test because you show output over time until the batteries discharge. By definition, that's a runtime test (with batteries at different starting temps). In fact from the first post... "Runtime is counted from the first reading to 50%." If all you cared about was initial performance, a plot of 0-10 minutes would have been much more illustrative of what you wanted to show/know.

And I'm not trying to be a jerk, it's just that you're putting down people with "read the entire thread" comments or banghead emoticons who are making valid interpretations about the data presented and the subsequent analysis of what is "best". I think it would have been helpful had the initial post explained that you were just investigating battery performance from T + 30 seconds until output levels off or just a straight 30 minute window with the graphs only showing that portion. That would have minimized any talk of runtime and focused solely on the initial performance.

It's great work and a useful test. I think there's one thing we can ALL agree on... alkalines really stink even more when they're cold than they normally do.

[Egsise]

The point of the freezer test is to see how low the output is before the batterys internal resistance, and flashlights driver circuit heats up the batteries.

I agree with you that my freezer test is good only for the first few minutes, but since I didn't find any similar tests this is better than nothing.

Whateva.

I need to find standard Duracell alkalines to test, but I can't find anything else than Duracell plus or M3 cells.

[Burgess]

to Egsise --


Thank you for your time and effort in performing these tests.



_

[DynoMoHum]

Very interesting testing. My interests are mostly related to batteries in general as opposed to flashlights and only recently decided that I'd start investing in LSD batteries. My previous thinking and needs were that LSD had little value to me personally since I typically want my rechargeable batteries for use in camera flash and had always figured I'd be willing to charge my cells fresh just prior to use most of the time and would rather have the extra storage capacity of non LSD NiMH. But anyway recently I decided to give the LSD cells a shot , since I need to buy some new cells.

Now, being a very curious person... I can't help but wonder how it is that the Eneloops do so much better in your tests then do the other cells you've tested. My curiosity is extended even further since you seem to imply that in similar tests of AAA Eneloop testing by this other Finish fellow that they didn't work very well. Since I can't read or understand the test results of that other person, I'm really not entirely sure just how bad things were in those tests

Either way, I wonder how much if any of your tests are related to the flash light electronics and such as opposed to the batteries themselves. I'd really like to see similar tests where ONLY the batteries themselves are tested for their ability to function at various temperatures. I guess I'll have to do some digging or do some of my own testing.

I understand how much effort it takes to do these kinds of tests and I greatly appreciate your doing them and posting the results here... You've certainly peaked my curiosity about cold weather performance. I don't need this kind of performance hardly ever, but if/when I should ever find myself stuck in some cold dark environment it sure would be nice to have some chance of shedding some light on things. Thanks for your efforts in this area.