Batteries for UltraFire AA and UltraFire Mini AAA

As many of you know, these are nice little units for the price, but they have a flaw: their regulator circuit requires 1.2-1.3 volts to start. In fact, the light produces output on as low as 0.2 or 0.3 volts, but only after it has started.

Unfortunately, this means that NiMH batteries--at least full-length ones--are out. Within a few minutes of use, they've lost enough charge that they can't be started.

Alkaline batteries are a bit better since they start at >1.5 volts, but they have enough internal resistance that it's not long before they can't start either.

Now, some people have suggested that the 10440 and 14500 LiIon batteries are a good fit. And indeed it seems so: the circuits themselves run fine at 3.6 volts, and in fact the circuitry seems more efficient at this voltage.

The problem is that the only places I can find these batteries are AW's sales here on the forums, and from Chinese sites like Fifth Unit and Quality China Goods. But because I'm going on my camping trip in a week, I need these batteries soon! The Chinese sites at least take a couple weeks to ship.

So--are there any alternatives? I was thinking about Energizer's 1.5 V lithium primaries. Maybe they have lower internal resistance and maintain their voltage better? I really don't know. Are there any 3.6 V lithium primaries that are AA or AAA size?

Ideally I could just find the 10440's and 14500's locally, but I don't think that's gonna happen. Maybe I could use 1/2 size NiMH AAA and AA's? I could probably get a hold of those more easily, but I'm not sure. Plus the runtime would be way reduced.

Any other ideas? Thanks!

-Scott

Aha! I just found that a CRV3 battery can be split into two AA-size 3V batteries. That would do the trick for the UltraFire AA, at least if I can find them for less than the $11 the grocery store charges.

Any similar tricks for AAA's?

-Scott

Sounds like L91 lithium AA's (around 1.7v) should work fine. They're fairly easy to find in brick/mortar stores.

Cool, it looks like L91 is the product number for the Energizer AA Lithiums. And when I googled the number, I found a nice datasheet:
http://data.energizer.com/PDFs/l91.pdf#search="L91 "

Looks like the discharge curve stays reasonably flat until it hits 1.3V, where it drops off quickly. Which is just perfect for this application. Thanks!

-Scott

Scott_C said:

As many of you know, these are nice little units for the price, but they have a flaw: their regulator circuit requires 1.2-1.3 volts to start. In fact, the light produces output on as low as 0.2 or 0.3 volts, but only after it has started.

Unfortunately, this means that NiMH batteries--at least full-length ones--are out. Within a few minutes of use, they've lost enough charge that they can't be started.


-Scott

Click to expand...

Ultrafire UF 601A 2xAA's...

A few minutes ?, you need some new nimh's.

I found it was at least an hour before I had nimh startup problems .

Total RT on 2,000 mah nimh's was 3.5 hours !

I can't speak for the AAA's though.

Have you tested the statup voltage on your UltraFire? I found that mine was a fairly consistent 1.3. That's above nominal voltage for NiMH, and about where the most are after the initial "peak" voltage burns off (doesn't take long).

Maybe you got lucky with your circuit, or you have a slightly different version?

No way would I get 3.5 hours of runtime on a standard NiMH, at least not with reasonable brightness. It's pulling way over an amp at these voltages. It peaks at 2 amps at 1.6V.

-Scott

Some power numbers I generated:

I had to stop the AAA test at 2.8V as it was getting very hot. Based on this, I'm guessing that a 10440 is actually a bad idea.

As you can see, the AA light peaks at 3.44 W and the AAA at 2.81. Pretty impressive! I hope the regulator is converting this power at reasonable efficiency. I don't have a good way of testing that.

In any case, it's clear that a standard NiMH might last 2 hours at reasonable brightness, but not much more. And as soon as it dips below 1.3 V, then you can't restart it (but it will produce light as long as you don't turn it off).

Looks like the L91's and L92's are the ones to get. Good looking discharge curve, and very high energy density. Too bad they aren't rechargable. On the AA model a 14500 might work well, since the regulator seems to do its job at not letting power increase indefinitely, but on the AAA I don't think that's viable.

-Scott

Scott_C,

I guess you measured UF-WF-602A1 and used a regulated power supply.
Using a regulated ps is good for repetitious accuracy , but different from real world using batteries .

Both my UF-WF-602A1 and HAIII version B-18W measrued 0,29A with fully charged AW14500 cells. The voltage under load is ~3,9V.

I agree.

In my real world, nimh current UF601A is 680ma at startup, obviously reducing as output declines.

3.5 hours is from max brightness to my baseline illumination for all my tests, approx equal to 12 led Ghost 2 initial output.

UF 602A1, single AA, brilliant torch, startup current with nimh is 1,170ma, RT 1.8 hours, same baseline.

The voltage, and current reduces as time goes on, PSU supplied volts cannot equate with time, when using batteries.

HiltiHome said:

Scott_C,

I guess you measured UF-WF-602A1 and used a regulated power supply.
Using a regulated ps is good for repetitious accuracy , but different from real world using batteries .

Both my UF-WF-602A1 and HAIII version B-18W measrued 0,29A with fully charged AW14500 cells. The voltage under load is ~3,9V.

Click to expand...

Yep, that's right. I'm using the WF-602A1 and the WF-602C on a regulated bench supply.

I guess your results are fairly consistent with what I saw--the current started dropping fast above 3 volts. I don't know if it would have hit 0.29A, but it might have gone as low as 0.5A. That's a pretty big difference but not too extreme.

I agree that a bench power supply is different from the real world, but I measure a given current and voltage that my batteries are running at, it seems like I should be able to match that with a PSU. Unless the numbers are being reported incorrectly--possible, I suppose, given the kind of regulators these flashlights use.

In any case, it's clear that for whatever reason, NiMH are not cutting it for me. I'm going to pick up some L91's and L92's and see how they do.

-Scott

abvidledUK said:

UF 602A1, single AA, brilliant torch, startup current with nimh is 1,170ma, RT 1.8 hours, same baseline.

The voltage, and current reduces as time goes on, PSU supplied volts cannot equate with time, when using batteries.

Click to expand...

Well, that seems very consistent with my measured numbers: your NiMH batteries are going to be in the 1.2-1.4 V range, and there I measured between 0.99 and 1.45 A current. So your measured 1.17 A is right in line for that range (especially assuming some voltage drop), and 1.8 hours total runtime seems reasonable for that current on a decent NiMH. It's just that for me, I couldn't start it up on less than 1.3 V. I've found this to be true for both batteries and my bench supply.

-Scott

Scott,

Battries have way higher dynamic internal resistance than bench psu, also the circuit does not draw current like a resistor, but pulsed current.

According to my experience, there is a major difference between your setup and battery operated lights.

Please measure current draw with psu at voltage over 3,6V and report.

Thanks
Heinz

My Ultrafire AAA runs great on a 10440 3.7 volt cell, a real firecracker, it does get hot, but I let mine run from full charge to 2 volts without any problems. Back in the real world... for a camping trip I don't think any AAA light is a very good idea. Either dim or poor run time.

My ultrafire AAA draws .45 amps from a fully charged LiIon AAA cell (about 4.1-4.2 volts). I get a very bright 15-20 min and another 10 min of dimmer. I was afraid to run it below 2 volts for fear of ruining the unprotected LiIon cell. Not very practical for a camping trip.

Holland

stjohnh said:

Not very practical for a camping trip.

Click to expand...

These lights are not my only ones . These are essentially emergency lights; something I can stuff into a pocket and have handy in case I need it quickly. Or give to a friend that forgot his. I have other, more powerful lights for general use.

Thanks for confirming that 10440's work, though. I realized that I had forgotten an extremely important factor: heatsinking. When doing my testing, I had separated the head from the rest of the body. But the aluminum body, in combination with the battery, provides a nice heat sink. So maybe in practice it won't be as bad as it seemed on the bench.

-Scott

HiltiHome said:

Battries have way higher dynamic internal resistance than bench psu, also the circuit does not draw current like a resistor, but pulsed current.

Click to expand...

Yep, that's certainly true. Hmm, maybe I should hook it up to an o-scope? I could put an 0.33 ohm resistor in line, and measure the voltage across it to see how the current varies. Anyhow, here are the numbers I measured:

WF-602A1:

You can see that the current is lower than my previous numbers. I believe that previously, the unit was getting quite hot at that point and affecting the results. For these numbers, I measured it in "pulsed" fashion, so the unit stayed cold the whole time. I believe they're more reliable.

-Scott

I use the acutal Ultrafire branded 3.6v AA batteries and charger in my WF-602A1. I ran several tests with this light vs. my Fenix L2P/L1P Body with the exact same batteries and found the regulation and runtime on the WF-602A1 to be superior. I am very impressed with this light and think more people should take a look at it.

Shao

You can get some li ons from lighthound in texas,he usally get them to you in 2 or 3 days!!! I ordered fri and had the m mon!

I couldn't find any in AA size...
Duh.. nevermind 14500...

Shao

The 14500 and 10440 cells I ordered from FifthUnit arrived--I'll try to test them tonight. The L91 and L92 lithiums worked very well on my camping trip, though. A bit expensive, but the long shelf life is a nice bonus.

-Scott

How long was your total purchase to shipped time from 5thunit? been thinking about getting some things from there...