Charge your battery as often as possible-revolutionary, comparative, numeric results!

[Guest Menneisyys]

I’ve, in all Windows Mobile forums, very often posted on the fact that Lithium-Ion / Polymer batteries should always be recharged whenever possible. That is, you should put your handheld / handset on the charger whenever possible. NEVER EVER wait before the handheld shuts down because even after only 4 months, your battery will only hold about a third (!) of its original charge holding capacity and, as opposed to what several people say, putting the device on the charger only once a day (or two-three days with devices with better battery life – mostly, TI OMAP-based handsets) on a regular basis (not waiting for the battery to completely deplete) will also have an adverse effect on the battery life.

Now, Groucho's pretty cool Nokia N95 blog has published a remarkable article, which is of extreme interest to anyone having a device with a Lithium Ion or Polymer-based rechargeable battery (not only Nokia N95 users – again, this info applies to ALL current Windows Mobile handsets, and a lot more other gadgets like digital cameras etc.). The figures he shows certainly prove I’ve been right when I recommended recharging your device whenever possible.

Groucho has purchased three N95’s (again and again, don’t think this discussion only applies to the N95 – it does apply to ANY device out there!) at the same time and used them in a pretty similar way. He, however, made absolutely sure he charged the device in three consistent (not mixing up the devices) and radically different ways: the first was charged only once a day (every night), not during the day; the second was charged when it has completely shut down because of the battery’s going flat and the third was almost always on charger whenever possible. The difference in the battery capacity, based on the different ways of recharging, turned out to be astonishing: the second battery (which was always let go flat, and only after that was it put on a charger) could only hold slightly more than a third of the battery that was always on charger whenever possible. The latter battery, in addition, fared considerably (about 16%) better than the battery only regularly (every night) recharged.

Note that the blog post has a mistake (which has no effect on the credibility of the results): it states they will have exactly the same charge holding capabilities throughout their life, and, suddenly, they die (“They have a finite shelf life which starts from the day they are manufactured and their condition from then degrades to the point where they fail! This is not to be confused with their charge holding abilities which are very good for their full life! When they go they just go!”). This is, of course, incorrect: their capacity will degrade over time. Just an example: my HP iPAQ 2210, which I bought right after it was released in Europe, at first, shut down at 2-3% battery level. After about two years, this raised to about 20% and, after three years, this was already around 35-38%. This also shows – along with a multitude of other examples – that these batteries do lose capacity over time. Also note that, as I’ve also recommended several times, Battery University has a decent article on all these questions – except for, of course, real-world results like the above.

Verdict: ignore if somebody says the opposite. Recharge your handheld whenever possible.

[Guest Monolithix]

Hmm, few things i'm not sure about with this article. First of which is his experimental method. There this is only one set of data, where are the repeats and control? Also, consumer batteries are notoriously wide-ranging in performance, even from the same batch. Each battery is made up from a set of individual cells, and each cell will have slightly different electrical ratings which have an adverse affect on performance when combined.

Secondly i don't think he's done much research into the performance of Li-Ion for use in batteries. I have a feeling that over the short term, his results may *possibly* be accurate, however over a longer period (eg that expected of the battery, 1-2years maybe?) the "higher" performing battery will quickly loose its capacity. There are two important design aspects to consider when building a battery. Firstly the "depth of discharge" (DoD), eg how much of the stored charge you use before recharging, and secondly the number of recharge cycles can be performed at this DoD. For Li-Ion you ideally want to have a low DoD, maybe 20%?, to enable the maximum number of recharge cycles. Ah, i hear you say, isnt that the conditions better performing battery was subjected to? Well, er, no. Due to precisely that, it's a *battery*, a collection of cells. So the constant charging and minimal self-powered use means its likely only a few of the cells are actually being used, to a lower DoD than the rest of the cells in the battery. So they'll wear out more quickly and the whole battery will perform far worse much quicker.

The overnight charge battery will probably perform the best in the long term imho, a reasonable DoD with regular recharging meaning you get a decent life and long-term use.

Furthermore, Li-Ion batteries have control circuitry built in to monitor and manage all this, in particular charging rates. Li-Ion is great but it has a nasty habit of exploding if you over charge it (all those Sony laptop batteries that exploded earlier this year? *that's* actually why they did it!). So its difficult to "mistreat" them, but still not impossible ;)

[Guest Menneisyys]

Firstly the "depth of discharge" (DoD), eg how much of the stored charge you use before recharging, and secondly the number of recharge cycles can be performed at this DoD. For Li-Ion you ideally want to have a low DoD, maybe 20%?, to enable the maximum number of recharge cycles.

Yes, or even smaller DoD. BTW, some batteries don't even let charging "kick in" over 92-95% charge level; for example, the Thinkpad notebook batteries. This means even if you keep your mobile device (this, in here, also includes notebooks) on the charger, it still won't charge the battery unless they're discharged under this threshold. (Of course, notebook batteries are pretty hard to compare to mobile phone / PDA bateries.)

What I haven't widely discussed (only mentioned when referring to the TI OMAP-based WM models) in the article is the different battery life of these models. While a, say, HTC Wizard has stellar battery life and sometimes only require one charge a week, this can't be stated of the Nokia N95 (or most Intel Xscale / Qualcomm / Samsung Windows Mobile phones), which, in cases, easily lose half of their charge a day even with moderate usage. That is, with a TI OMAP device (unless you use it a lot, run apps that use its CPU at 100% etc), it's not as important to be kept on the charger whenever possible as with, say, the pretty power-hungry Nokia N95, which only has about a three days' battery life with VERY moderate use.

Secondly i don't think he's done much research into the performance of Li-Ion for use in batteries. I have a feeling that over the short term, his results may *possibly* be accurate, however over a longer period (eg that expected of the battery, 1-2years maybe?) the "higher" performing battery will quickly loose its capacity.

Let me disagree. While such a situation is, theoretically, possible, in practice, Li-Ion batteries are known for gradually losing their capacity, depending on both time, full (!) recharge cycles and the DoD (all three having an adversive effect on the charge hold). In practice, however, it has turned out that the aging of the Li-Ion batteries doesn't tend to quickly accelerate over time. Just take an example of my HP IPAQ 2210 battery, which lost about 11-13% capacity each year, meaning only having about 62-63% of the original capacity after exactly three years. Aging (that is, the battery losing its capacity) didn't accelerate at all - it was pretty much a constant value (about 12% each year) during its life span. (BTW, I've also kept the battery on charger whenever I could.)

Furthermore, Li-Ion batteries have control circuitry built in to monitor and manage all this, in particular charging rates. Li-Ion is great but it has a nasty habit of exploding if you over charge it (all those Sony laptop batteries that exploded earlier this year? *that's* actually why they did it!). So its difficult to "mistreat" them, but still not impossible ;)

Fortunately(?), only third-party batteries suffer from this problem - not factory ones. I've never run into a factory PDA / phone battery being damaged from constant (re)charging. Notebook batteries are a different matter because they have to endure orders of magnitude more wear-and-tear (Li-Ion batteries hate high Amperages, which definitely is the case particularly in notebooks not having a mobile CPU); hence the burns and problems widely known.

BTW, all (factory; that is, high-quality) Li-Ion batteries have always had charging protection circuitry in them. That is, unless the circuitry itself goes wrong (never happened on me - again, I'm only speaking of mobile phone / PDA batteries, not notebook ones, which are an entirely different class), it's impossible to destroy a Li-Ion battery by just leaving it on the charger - it won't overcharge.

[Guest Menneisyys]

There is some really interesting discussion of this article at MoDaCo, HowardForums, FirstLoox, PPCT, XDA-Developers - well worth checking out for more info / discussion.