AA Limitations

[divine]

It seems like lights with the AA form factor cannot drive an LED to it's full capability.

AA nimh's (Eneloop) are capable of 4A of current, but AA alkalines can only give out ~ 1.2A.

So, a nimh is able to handle enough power to drive a Q5 at happy levels, while an Alkaline can only drive it at ~100lm levels.

Do the flashlight manufacturers have to design it as if people are going to run alkalines in it so people don't damage alkalines in their light and ruin it or return it under warranty?

Am I missing something about the driver circuits, or am I correct about my thinking? (The battery designs making the manufacturers lower the output based on the possible batteries that might be put in the light.)

Thanks!

 

[Mr Happy]

I think this may be true of low end consumer lights.

I can't say I've seen many lights where the instructions say to use NiMH and not alkaline, although a few of the more expensive lights aimed at enthusiasts show output/runtime graphs demonstrating much better performance on AA lithium or NiMH compared to dismal results on alkaline (which amounts to a recommendation not to use alkaline for all practical purposes).

As a matter of interest, I don't think CR123A can really do 4 A either. The voltage sags like crazy if you try to draw that much current from them.

My personal preference for a form factor would be 4 AA (NiMH) cells in 2 x 2 formation giving a slim rectangular handle and a 4.8 V supply with plenty of current. Sadly, very few lights are made in this form factor.

 

[n3eg]

My personal preference for a form factor would be 4 AA (NiMH) cells in 2 x 2 formation giving a slim rectangular handle ...Sadly, very few lights are made in this form factor.

That's because people won't buy them because they can't figure out how to unscrew the tailcap.

 

[metlarules]

I think this may be true of low end consumer lights.

I can't say I've seen many lights where the instructions say to use NiMH and not alkaline, although a few of the more expensive lights aimed at enthusiasts show output/runtime graphs demonstrating much better performance on AA lithium or NiMH compared to dismal results on alkaline (which amounts to a recommendation not to use alkaline for all practical purposes).

As a matter of interest, I don't think CR123A can really do 4 A either. The voltage sags like crazy if you try to draw that much current from them.

My personal preference for a form factor would be 4 AA (NiMH) cells in 2 x 2 formation giving a slim rectangular handle and a 4.8 V supply with plenty of current. Sadly, very few lights are made in this form factor.

It would be sweet if Princeton Tec would make an aluminum version of their pt40.

 

[TigerhawkT3]

Many manufacturers recommend the opposite: only running alkalines. Sometimes this is just for safety certification purposes, but several lights rely on alks being really weak. These lights, mostly in 3xAAA, rely on the internal resistance of the battery to limit current.

2x2 AA is indeed a comfortable arrangement - one of the many reasons to own a ProPoly.

2.5A is pretty much the limit for CR123As, as seen in some of SF's high-powered incans.

 

[divine]

Mr. Happy,
A primary AA battery will do 1.2 amps at 1.5 volts, so it can deliver approximately 1.8 watts.

A primary CR123A will do say... 1.2 amps at 3 volts, so 3.6 watts or or so.


I was asked why a single cell CR123A light can produce so much more light than a single cell AA light. When I was thinking of a response... both batteries can deliver approximately the same power for the same amount of time, but the current capability of the primary cells is different. (The weaker cells, which the manufacturer has to design for... worst case.)

See... if you design for 2 or 3 amps out of a 1.2 volt cell, you put in a 1.5 volt cell, when the current starts coming out, the voltage is going to dip and you will get much less light out.

Or so I think. :thinking:

 

[Marduke]

It's best to just ignore alkaline cells, they are good for practically nothing these days. NiMH cells are FAR superior, and a single L91 even packs more energy than a CR123 cell (along with a longer shelf life, and the ability to be used on common appliances)

 

[dfischer]

Divine:

I'm a real rookie so take this with a grain of salt. but:

While these drivers seem to deliver reduced amperage output @ lower voltages, there are a number of probable reasons. But who cares why? It's just a sorry fact. At present if you want max light output you need to be @ or near 3V. From there the drivers seem to be able to deliver consistent amperage. More of a regulator I suppose.

Anyway, if you want to use n X format then it seems the choices are to run 1x2 (parallel) of the lion hi-voltage stuff or run 3x1 (series) of the nimh Low V stuff.

I kind of wonder if a 25500 in a C cell light would leave enough space to run ribbon cable alongside it. I don't think it would be too hard to make a 2xC light accept two 25500 in series that way. That and a new optic in a lowes Creee is a project I've in mind.

Or, the DX C3's with an extra adapter with 3xAA's will drive nice is nice as well.

Both should have the same driver output and ability to drive an LED to pretty high levels. The Cx2 in series LI should have about twice the run-time. But I've not seen any protected ones yet..

 

[UnknownVT]

Am I missing something about the driver circuits, or am I correct about my thinking? (The battery designs making the manufacturers lower the output based on the possible batteries that might be put in the light.)

This was definitely true for incandescents -
eg: the "ultra bright" xenon bulb 2x or 3x CR123 lights like the now legendary SureFire 6P and 9P - there was no way AA alkalines could match that in any convenient form.

However with the advent of the more recent higher efficiency LEDs like the Cree XR-E Q5, Rebel 100's etc - the AA alkaline seems to do as well - but usually in the 2x AA factor -

But a 1x AA light still has exactly the limitation that you alluded to.

Here's an example - the Fenix "L" Digital series - the Fenix L1D and L2D (as well as P2D) heads are the same/interchangeable. Let's take the brightest Q5 version -

L1D-Q5
General Mode: 12 lumens (25hrs) -> 53 lumens (5hrs) -> 107 lumens (2hrs) -> SOS
Turbo Mode: 120 lumens (1.5hrs) -> Strobe

L2D-Q5 -
General Mode: 12 lumens (55hrs) -> 53 lumens (10.5hrs) -> 107 lumens (4hrs) -> SOS
Turbo Mode: 180 lumens (2.4hrs) -> Strobe

P2D-Q5 (single CR123)
General Mode: 12 lumens (30hrs) -> 55 lumens (5.5hrs) -> 107 lumens (2hrs) -> SOS
Turbo Mode: 180 lumens (1hrs) -> Strobe

Notice the brightest Turbo mode - the L1D-Q5 delivers a claimed 120 lumens, whereas exactly the same head/LED manages a claimed 180 lumens on 2x AA in the L2D-Q5 or 1x CR123 in the P2D-Q5.

Again as you point out in your later post #6 - although the 1x CR123 P2D-Q5 delivers the same max brightness as the 2x AA L2D-Q5 - it has less than half the runtime - so it was the AA's limitation in ability to deliver the required amount of current - at the required voltage that prevented the 1x AA L1D-Q5 of being as bright on Max as either the 2xAA L2D or 1x CR123 P2D - and the 1x CR123 P2D of being able to be as bright on Max as the 2X AA L2D - but not having as a long a runtime.

 

[mdocod]

There are several reasons that it isn't practical to make a 1AA light powered light that drives the LED at maximum potential (~4-5W range)

From a ~1.2V cell, inlet current would need to be pretty high, 4-5 amps or more, this is not a problem for many cells out there, but is a major problem for the design of the boost circuit, as the inlet side needs to be able to handle 5 amps, and at the same time, needs to have VERY low resistance doing it because you only have ~1.2V to work with. With such a low source voltage, even very small amounts of resistance will drop the available voltage substantially, which would result in even greater current requirements to overcome the voltage sag, which would in turn, sag the voltage of the cell even further. It would not be surprising to me if in the end, the efficiency of such a system would be as low as 65%, so by the time it's all said and done, you're actually pulling 6+ amps to drive a cree at ~4.5W.

High amperage requires heavy gage wire runs and larger inductors on the boost circuit, possibly so large that it wouldn't even fit in the size of a AA bodied light.

Have a look at the Jetbeam 1AA light... on a 14500 3.7V it CAN push that cree right up there with some of the best performers, but it can do this because it's not having to deal with such a low voltage high current "problem" when it has a 3.7V cell in it. the performs drops in ~half roughly when a regular 1.2V cell is used.

A single CR123 can do it also, again, because it has more voltage to work with, and current requirements are more appropriately balanced with the available voltage.

 

[kaichu dento]

That's because people won't buy them because they can't figure out how to unscrew the tailcap.

That's only for the first 10 seconds, then they realize that the head comes off.

My favorite light form used to be the flourescent orange 4xAA Buck Lights, but I always carried them in my jacket pockets due to the size.

 

[Meterman]

Mdocod has hit the nail right on the head.

With low voltage and thus high amperage the inevitable losses in body, tailcap, spring, switch, and mainly the connections between them will be very high, I think, this could rise to a few tenths of a volt easily.

It's not easy to measure all these resistances correctly due to the difficulty of connecting the necessary Kelvin Clips, but I'm trying to find a different solution here. Ideas are "under construction" already.


Wulf

 

[TorchBoy]

Re: AA limitations

Mr. Happy,
A primary AA battery will do 1.2 amps at 1.5 volts, so it can deliver approximately 1.8 watts.

You're not talking about alkalines now, or even lithium primaries, as Silverfox's 1 amp tests show. They all drop almost immediately to 1.4 volts, and it's downhill from there. 1.2 amps would be even harder for them.

There are several reasons that it isn't practical to make a 1AA light powered light ...

Sorry, but I can't even imagine what a light powered light is, unless you're talking solar panels.

 

[zipplet]

Re: AA limitations

He refers to a light that is powered by another 1xAA light. For example, the body tube accepts a Fenix L1D, and it uses some spectacularly efficient solar panel to generate power to run an HID lamp.

Edit: Ofcourse, that would break the laws of physics.

 

[divine]

I forgot to think about the voltage drop!

Well done, thanks guys.

 

[Curious_character]

Mr. Happy,
A primary AA battery will do 1.2 amps at 1.5 volts, so it can deliver approximately 1.8 watts. . .

Not for more than a couple of seconds. See the second graph here.

c_c

 

[zipplet]

For what it's worth, I tried flash amping an AAA LSD cell ("Vapex instant" brand) and got 10.8 amps after 1 second. Impressive.

 

[Meterman]

I've burdened an eneloop AAA (as little as 800mAh only!) with 10A at my e-Load.
The cell was connected to the e-Load via my self-built battery-adapter (measured 9.7mΩ, let's say 10mΩ), so at 10A the adapter costs me 0.1V loss. This must be added to the values in the diagram to get the real voltage of the cell.

Load was switched on at 23s, voltage begins to diminish at about 43s.

Current (left scale) is shown in green, so voltage (right scale) is blue.


Thumbs up for eneloop!


Wulf