AAs vs proprietry Li-ions

altis

Newly Enlightened
Joined
Oct 26, 2005
Messages
124
Location
Warrington, UK
Although I persist in trying to buy stuff that uses AAs I keep being told that I should 'get real' and use Li-ions instead. My hatred of the things started with a laptop that used a pair of camcorder batteries. These didn't seem to last very long and were very expensive to replace. With digital cameras, in particular, it's getting more difficult to find new models that don't use some proprietary Li-ion. The manufacturers are cottoning on to this money spinner. Is it time to bite the bullet?

Here's my take:

Availability
The AA format must be the most common battery shape on the planet. Replacements can be found almost everywhere. Many new cameras come with new Li-ion batteries and, for a while, the manufacturer is the only source. Later, as the clone makers bring at compatibles, there are other sources but there are so many models that most retail outlets are unlikely to stock but a few.

Price
Since the AA format is a common standard, the battery manufacturers are forced to compete on price. Proprietary batteries are much more expensive.

Spares Management
Looking after spares for a variety of battery-powered devices is made much easier if they all use the same battery. In established products AAs are probably the most common.

New Developments
Engineers are always developing new battery chemistries with better characteristics. The capacity of proprietary Li-ions hardly, if ever, changes but newer, better AAs are available from time to time.

Flexibility
Many different chemistries are available in an AA package. This allows the user to tailor the battery to the environment eg: rechargeable around town but lithium primaries while away from the mains. If the battery compartment is carefully designed it is also possible to use a CR-V3 (or a rechargeable version) in place of two AAs. With a proprietary battery you are stuck with what the manufacturer provides.

Self-Discharge Rate
This is the Achilles heel of ordinary NiMh cells but now we have Eneloops (and cousins) that hold around 85% after a year. This works out at 1.3% per month. Various rates between 1% and 5% per month are quoted for Li-ions (with the highest rate being most common).

Energy Density
This is difficult to calculate accurately from datasheets. Really, we need the area under the discharge curve under the same conditions but this is difficult to evaluate without actually measuring it. Perhaps the easiest way is to compare cells of the same size but with different chemistries. Using the data from GP Batteries, their '2700' high-capacity NiMh cells typically hold around 3Wh (calculated from 1.15V * 2600mAh). Their 14500 Li-ion cells have a capacity of around 2.5Wh (calculated from 3.6V * 700mAh). For the CR-V3 (double AA) it's around 4.1Wh (2.5V * 3300mAh / 2). For Energizer's e2 lithiums it's 4.5Wh (1.5V * 3000mAh) and for Sanyo's Eneloops it's around 2.5Wh (1.25 * 2000mAh). The non-rechargables have the highest energy density but, of the rechargeable, the high-capacity NiMh is best.

Of course, lithium-based cells are very light so, by weight, they have a better energy density. Energizer's e2s weigh only 14.5g so the gravimetric density is 0.31WH/g. For the CR-V3 (38g / 2) it's 0.22Wh/g and for the 14500 (19g) it's 0.13 Wh/g. For the 2700 (31.5g) it's 0.095 Wh/g and for the Eneloop (27g) it's 0.093 Wh/g. Advantage lithium - but the non-rechargables are by far the best.

Discharge Rate
Lithium-based cells tend to be rated conservatively to stop them overheating. The maximum recommended discharge current of GP's 14500 is 700mA whereas the 2700 NiMh cell is rated for 7.5A. Being able to draw lots of current is useful in applications like flashguns where this helps to reduce the cycle time.

Charge Rate
Likewise, lithium-based cells are not comfortable with fast charging. The fastest chargers seem to take about 2 hours but NiMhs will tolerate being charged in 15 minutes. Sure, this will reduce their cycle life if done repeatedly but not if you only do this from time to time.

Lifetime
Lithium-based primary cells have a tremendous shelf life but can only be used once. Li-ion rechargeables are limited by time since manufacture. We used to get 2 years but newer models last 5-7 years. NiMh cells are limited by number of cycles. Sanyo quote up to 1000 cycles for their Eneloops - equal to once a week for 20 years.

Safety
Batteries containing lithium have a tendency, albeit slight, to overheat and catch fire. There are many examples to be found on the web. A 48-pack of Energizer e2 lithium primaries that I recently purchased had a warning on the outside to say it was unauthorized for air transport. All spare lithium batteries now have to be carried in insulating containers on a plane. Alkaline and NiMh cells do not have such problems.

Datasheets:
GP 2700 NiMh
GP 14500 Li-ion
GP CR-V3 photo lithium
Energizer e2 lithium
Sanyo Eneloop
 
Last edited:
I think you should probably take into account discharge curve characteristics and voltage-drop under load, not to mention try to compare AA cells to the most commonly-used lithium-ion cell, the 18650. It seems to have panned out that the 18650 is a good-sized cell for a most torches, especially when one considers that nominal capacity is 3.7V*2.2Ah = 8.14Wh. That's over three times the energy of an AA cell in a nice lightweight package - yes, it's larger than a single AA cell, but not all of us want a light that's AA, 2AA, or even 3AA sized - a typical 18650 light fits my hand nicely. It seems to me that as lithium-ion cells get larger, the comparative energy-density increases relative to multiples of NiMH AA cells... as it should. More space for reactant and less space for casing, anode, cathode, etc.

You also have neglected the new generation of A123 and Emoli lithium cells - which are capable of repeated fast charges and extremely fast discharges without reducing cycle life.

I personally have made an effort to reduce my battery store to AAs and 18650s, so I'm definitely with you 100% on the positive aspects of LSD NiMH technology. But... I'm looking forward to further advances in lithium-ion technology which will hopefully make them safer, more powerful, and less expensive.
 
I think you should probably take into account discharge curve characteristics and voltage-drop under load...
I have!

In each case I looked for the 1C discharge curve and estimated the voltage at the mid-point so as to make my best guess at the area under the curve. This is why, in some cases, I've used voltages that are not the 'normal' ones.

Is the 18650 more representative of the prismatic cells used in camera batteries?

I don't know but lets take a look at it anyway. The 18650 (which, as far as I am aware, is a Li-ion only size) has a volume of around 17,000 mm3. For an AA it's around 7,700 mm3 or about 0.45 of the size. Using the 1C discharge curve on the GP datasheet I estimate a capacity of about 3.5V * 2200mAh = 7.7Wh. Scale this by the ratio of volumes and we get 3.5Wh which is less than the 4.5Wh of Energizer's e2. Yes, this is more than the 3Wh of the NiMh - but only just. And is this worth the disadvantages?

http://www.gpbatteries.com/pic/GP1865L220 TDS1.pdf
 
Last edited:
My main reason for switching back to Eneloops from Li-ions is that protected Li-ions cut off with no warning. You can either constantly recharge your Li-ions or you can get stuck in the dark with no warning. I'm not going to use unprotected Li-ions.
 
Excellent and thorough comparison. I'll stick with Eneloops (AA and AAA) and a good charger/analyzer. My future electronic device purchase decisions will be strongly based on whether or not the new device runs on AA or AAAs.

To complete this picture, we need a good, portable solar charger that can charge a bank of 8 AAs on one not-so-sunny day.
 
In many of the small point-and-shoot cameras, LiIon packs are much smaller than 2xAA. Really prismatic is the only way to go in those form factors, but for full-size cameras, where size/weight don't need to be limited at all costs, you do make a good argument for AAs -- those are especially nice for people who like to "standardize" and use the same cells for everything, like eneloops.

At home I have a couple canon cameras, and a canon DV camcorder, and luckily they ALL use the same type of battery, so I Was able to get some brand-X spares, and start a rotation of them.
 
Is the 18650 more representative of the prismatic cells used in camera batteries?
I'm not sure about camera batteries - I hate lithium-ion camera batteries, they're all expensive, all proprietary, and all annoying. I never buy cameras that don't run on AAs.

... I estimate a capacity of about 3.5V * 2200mAh = 7.7Wh. Scale this by the ratio of volumes and we get 3.5Wh which is less than the 4.5Wh of Energizer's e2. Yes, this is more than the 3Wh of the NiMh - but only just. And is this worth the disadvantages?

http://www.gpbatteries.com/pic/GP1865L220 TDS1.pdf

I'm not going to dispute your math, which seems to be right in line - my apologies for not noticing your compensation for voltage drop under a 1C load. However, your logic may be interpreted as flawed with respect for a target Wh capacity for any particular device. For instance, obtain the energy contained within an 18650 it takes approximately 3xAA according to your calculations. 3xAA cells take up quite a bit more space than a single 18650 which can vary based on how you arrange them... in a triangle, they take up about the same volume as a D-cell... which is quite a bit larger than an 18650. I can pocket-carry an 18650 light... can you pocket-carry a 3xAA light? EDIT: To clarify, I'm using your Wh figures for the Eneloops... because LSD is an important feature for most people of any power source.

Again, I'm not disputing your calculations, just the application of values to real life...
 
Last edited:
I'm not sure about camera batteries - I hate lithium-ion camera batteries, they're all expensive, all proprietary, and all annoying. I never buy cameras that don't run on AAs.
..


Of course, that statement eliminates Almost all of the highend cameras. Some Pentax DSLRs use 4 AAs, but all of the other DSLR cameras use proprietary batteries. And a lot of the upper end compacts use LiIons too. That's the only way to get enough power into the incredibly cramped body of these new cameras.

And I have a Canon EOS Rebel XT which uses LiIon and a Canon S5IS that uses Eneloop AAs. I also have 2 Canon film cameras that use Primary Lithium batteries. Two flashguns use AAs too.

Yeah, it's nice to just stock up on Eneloops but Canon, Nikon, Sony and Olympus have no intention of bringing out a AA powered DSLR. Leaves a couple of Pentaxes if that's what locks you in.

BTW, I have an external battery grip for my XT that either takes 2XNB2LH batteries or 6XAA NiMHs. 1000 shots on the NB2LHs and about 1500 with the AAs. And I can load both trays and be ready for a 30 second swap and 2500 pictures. Of course, 500 shots on each of 3X2 gig cards and 1000 on my 4 gig and if I'm using the 430EX flash I'll be swapping batteries in it about every 250 shots. Situations I doubt very seriously I'll ever be in. Just a bit of info.

And a collection of rather non-descript flashlights, including 2 X 3D M@gs, 3 X AA M@gs, 2 with Wally Werld drop-ins, a 3 X AA M@g LED, a Brinkman 3 watt LED, (it's nice) a Brinkman 2 X AA LED thaT is fairly bright for 5 years old, A Brookstone 3 X AAA LED Headlight that is very handy and seems nice and bright and a Brookstone 3 X AAA 1 watt with optic that is a nice (blue tint) light.

Gonna have to get some drop-ins for the 3Ds.
 
altis, all very good points. I'd like to try to set the record straight on a few things here, but many of your points are very good. Nothing against standardizing on AAs, I have increased my collection of NIMH cells to include 24 AA eneloops and absolutely love them.

PRICE:
If you factor in how long the typical NIMH cell lasts as far as cycle life, and how much capacity they loose through their cycle life, some initially expensive investments in a Li-Ion battery can actually break even or come close in the long run while delivering a more user friendly experience. Eneloops and other LSD cells are changing this with exceptional cycle life. But it was only "yesterday" that NIMH cells in most cases rarely had a useful life over 200 cycles. Many consumer brand NIMH cells like the 2500mAH energizers had useful cycle life around 10-50 cycles for most people before developing the most insane self discharge rates imaginable. (Making the cells useless for most applications).

NEW DEVELOPMENTS:
18650s are as common to laptops as AAs are to consumer devices. They are seeing constant revisions to improve capacity and safety. Pushing "2600mAH" label capacity now from some of the good ones like LG. And we are now seeing lithium iron phosphate, lithium nano iron phosphate, and lithium manganese oxide which are still in their infancy but are capable of delivering huge current while surviving hundreds or even thousands of cycles with minimal self discharge or voltage suppression issues. New tech is emerging that will continue to improve the capacity of these new safe chemistries. Now that these cells are popping up in power tools and cars *soon* we are going to see a LOT of advancements in the next decade with li-ion. My guess would be that Li-Ion will have more advancement and R&D than all other cell types combined in the next 10 years.

FLEXABILITY:
there are a number of 3rd party sources for replacement battery packs and chargers for all sorts of consumer devices, from laptops to video cameras. Some of them offer higher capacity and lower prices than their OEM offerings. But I agree that some Cameras out there use packs that are only available from 1 place in 1 size and with 1 charger. I try to avoid that kind of stuff :)

SELF DISCHARGE:
Silverfox has done a 1 year self-discharge test on a bare Li-Ion cell and came up with 95% remaining capacity. NIMH is still not quite as good as Li-Ion as far as self discharge is concerned. The other issue is that even LSD NIMH cells do most of their self-discharging in the first few days and weeks off of the charger, then the rate of self discharge slows down, whereas Li-Ion holds very close to "fresh off the charger" capacity for weeks after being charged. So you can't just look at discharge per year characteristics, but how that discharge takes place and when. Most people will use their device within a few days/weeks/months of a charge, in which case, Li-Ion will win hands down in the self-discharge arena.

ENERGY DENSITY:
comparing a 14500 size Li-Ion to AA size NIMH is very unfair, as 14500 Li-Ion cells are one of the smallest sectors of the Li-Ion market and are only developed to fit a few niche products. There is very little R&D going into that size. On the other hand, as you have conveniently pointed out, AAs are EVERYWHERE, which means they receive more R&D than any other NIMH cell size. It would be unfair to compare the NIMH cell that gets so much of the money spent on R&D to the LI-Ion cell that gets practically no attention. In order to fairly compare energy density of NIMH to Li-Ion you want to take AA NIMH vs 18650 size LI-Ion. The 2 most common cell sizes in their respective chemistries. At ~1C NIMH AAs range from around 2.1WH up to around 2.7WH (looking at the range of LSD up to the "best" high capacity cells available. At ~1C 18650 cells range from about 6WH to about 8WH (the range of ~2000mAH labeled 18650s up to the best of the best 2600mAH labeled cells). a NIMH AA cell is about 7.7 cm^3. Or about 0.273-0.351WH/cm^3. An 18650 cell is about 16.54 cm^3. Or about 0.363-0.484WH/cm^3. Now since the "ultra-high" capacity NIMH cells can't really be compared to Li-Ion for most uses because of their very high self-discharge rates, you are kind of forced to compare Li-Ion to LSD cells to be on the same playing field for other features. In which case the higher end 18650 cells are actually about 75% more energy dense than LSD NIMH cells. Comparing energy density of Lithium primary cells to rechargeable format cells is a moot point. The best primaries will always outclass secondary cells on energy density, whether you compare to Li-Ion or NIMH. Just FYI, a lithium primary AA delivers about 3.7WH into a ~0.3C load. I haven't seen a discharge graph of one at ~1C.

DISCHARGE RATES:
Again, I stress the importance of comparing AA NIMH to 18650 LI-ION. While NIMH can handle some very intense current demands, you have to also factor in that they suffer dramatically in the from of lost cycle life when subjected to repeated demanding loads. There are hobby grade NIMH cells out there designed to deliver up to ~20+C loads, but they don't last long, by comparison, Lithium nano-phosphate cells are delivering 2-20+C loads in power tools and lasting hundreds of cycles. A pack of 3 NIMH LSD eneloops in series delivering delivering a 4A load will last about 26 minutes and deliver about 6WH worth of juice. a LG 2400mAH 18650 will last 31 minutes into the same load, delivering 7WH. 4A is a pretty reasonable comparison point for a high drain consumer device, probably a lot higher than you will find in 90% of devices but within reason. There is a point to be made for the tolerance and safety of NIMH when stressed, it does not become unstable and dangerous when stressed repeatedly, standard Lithium CoBalt Oxide cells require monitoring of cell health when subjected to high loads frequently.

LIFETIME:
Most cycle life ratings are pretty optimistic, we don't have any proof that an LSD cell will survive 1000 real world cycles and still be worth beans, but we do have a lot of evidence that properly implemented Li-Ion cells can in fact deliver hundreds of usable cycles in the real world.
 
Although I persist in trying to buy stuff that uses AAs I keep being told that I should 'get real' and use Li-ions instead.
I don't know who told you to "get real", but it sounds like you have solid arguments with sticking with AA's. :thumbsup:

I only use li-ions in a couple of lights.
 
Nikon in fact makes an AA powered DSLR (the D300 with the AA accessory grip) and aftermarket AA packs are available for several other models. In consumer digicams, AA actually seems to be gaining rather than losing popularity. There's tons of good models.

I completely agree with the reasoning in the first post. What I -really- want is an AA powered cellular phone, for all the reasons mentioned above. THAT seems impossible to find, unlike the digicam situation.
 
I've been told similar, that decent stuff uses lithium ion. Most times I find their opinion is based on having one device that they ran a set of NiMH's from walmart in so they were disappointed in them.
I have standardized on AA eneloop cells for anything that I can.
 
Nikon in fact makes an AA powered DSLR (the D300 with the AA accessory grip) and aftermarket AA packs are available for several other models. In consumer digicams, AA actually seems to be gaining rather than losing popularity. There's tons of good models.


And as I said, I have the Battery Grip for my Canon. But Pentax is the only manufacturer that makes a DSLR that takes AAs in the camera body. Accesory grips are another issue and handy at times, bulky at others, heavy in still third situations.

And AAs are hanging on quite well in lower end cameras, not as well in higher end stuff.

And I would like my cell phone to take 4XAAAs. Smaller and lighter than AAs, yes, less mAh, but the same arguments gor for AAA as AA and they have even more to recommend them. I have 2 batteries for my phone and keep both charged. swap out when I think about it. It never takes more than about 2 hours to fully charge from dead though, and that is hooked to the powered USB hub on my puter. I have the AC adaptor, but rarely use it, and also have the car cord. So I really, at this point, have no need for a different power supply, but for my next phone, the option of AAAs would be nice. I'm not holding my breath.
 
Depends on what you call higher end. DSLR doesn't necessarily equal high end, there are low end and high end DSLR and low end and high end point and shoot. For example the Canon S3IS or S5IS is a high end point and shoot.
DSLR is a different market, the pros doing paid photography want the most power available in a package which lithium ion still has a slight edge over NiMH and that tiny bit counts if your getting paid for a photo shoot. They don't care if they have to replace the battery pack in a couple years or its super expensive, its just the cost of doing business.
I stick with AA for standardization purposes. Lets see we have three different digital cameras that take AA's, GPS, Scanner, CapShare, CB, flashlights, etc. Now imagine they all had a lithium ion battery so I'd have to have a charger plugged for all those and the rats nest of power strips and extension cords. it would take a completely separate pack to carry just the chargers. With everything running on AA's all I need is one good AA charger.
 
Some good discussion there guys - especially Mdocod.

Last night I remembered an advantage of the Li-ion that I forgot to mention. If the chemistry of the battery is fixed then the camera (or whatever) has a better chance of knowing its discharge characteristics and can have some attempt at a battery charge meter. Some Li-ion packs, although not many, have an integrated charge meter (Info-Lithium) for accurate tracking of capacity.


Back to energy density:
I have just measured the prismatic Li-ion in my mobile phone. This is 36x50x7mm and has a claimed capacity of 600mAh at 3.6V. That's 3.6V * 600mAh = 2.2Wh in 12,600 mm3. Scale this back to an AA's volume and we get 1.3Wh - considerably less than any of the other cells mentioned so far.

Received wisdom is that Li-ion cells have a higher energy density than NiMh. I believe that this is not always the case.
 
I don't know but lets take a look at it anyway. The 18650 (which, as far as I am aware, is a Li-ion only size) has a volume of around 17,000 mm3. For an AA it's around 7,700 mm3 or about 0.45 of the size. Using the 1C discharge curve on the GP datasheet I estimate a capacity of about 3.5V * 2200mAh = 7.7Wh. Scale this by the ratio of volumes and we get 3.5Wh which is less than the 4.5Wh of Energizer's e2. Yes, this is more than the 3Wh of the NiMh - but only just. And is this worth the disadvantages?
At least if you include tagged cells, there are certainly 18650 NiMH cells around, as well as various other cells very roughly the same size (4/3A, 7/5AF, etc)
Typically the high-end 18650 NiMHs have 4500mAh capacity, which equates to 5-5.5Wh.
 
Depends on what you call higher end. DSLR doesn't necessarily equal high end, there are low end and high end DSLR and low end and high end point and shoot. For example the Canon S3IS or S5IS is a high end point and shoot.
DSLR is a different market, the pros doing paid photography want the most power available in a package which lithium ion still has a slight edge over NiMH and that tiny bit counts if your getting paid for a photo shoot. They don't care if they have to replace the battery pack in a couple years or its super expensive, its just the cost of doing business.
.


S3IS (last years) and S5IS are higher end cameras and I have an S5IS. The G9 is Canon's top compact though and it's lithium. And about 90% to 95% of DSLRs are NOT sold to pros. I know i have a DSLR and I'm a long way from PRO. And I like that little NB2LH battery because it's way smaller than any equivalent in NiMH batteries, which would 6XAAA to give the smae voltage and 720 mAh capacity. And the BGE3 grip allows me to use 6XAA and adds significantly to the weight and bulk of the camera.

So, for using AAs, we can either have a much larger camera to have the space for 6 of them, add a $200 external battery grip, go to a Pentax which does not have the same capabilities, especially since the ones that use AAs are their bottom end, or stick to a next to the top compact from Canikon. I want the camera to do what I want and I don't want more bulk and weight than necessary.
 
I hate proprietary packs of any stripe, and especially one-offs rather than, say, the NB-2L or the NP-FMxx. Though, I do wish the aftermarket versions weren't so lousy. Big range between $20 and $70 to make a better cheaper pack.

Proprietary packs without charge monitors (like infolithium) suck, to put it mildly. You get a, generally useless, "low batt" indication and that's it.

The only excuse to have a proprietary, no gas gauge pack is, alas, the main factor in my last two gadgets, a Sanyo SCH-8400 phone, and a Casio Z77. I don't think even AAAs could have been shoehorned into their battery compartments, and size was very important. The Sony W1 it replaced takes better photos and uses 2xAA, but is too thick, and no MPEG4 video.

One success, though. I leave my iPod mini at home in favor of a Creative Labs Muvo v100. 2GB, plays Audible.com books, loads like a flash drive, and... runs over 10 hours on a single NiMH AAA.

Oh, the other would be the Fenix L0D. It _was_ 10440 powered until I found out how unsafe a 3C rate is.

So, rly, I'm beginning to standardize on AAAs.

(The phone, BTW, will run 4 days on one charge with light use)
 
Top