Best practice when charging a Li-Ion battery?

[Belfaborac]

I've had a look around for an answer to this without coming up with anything, but maybe there's a geek here who knows....

I'm aware of the main differences between charging a Li-Ion and a Nickel-based battery, but there's one thing I'm unclear about. With a NiCad or NiMh a trickle charge (typically 0.2A) is *generally* better than a "nuclear blast" (a 2A current for instance) of the type generally delivered by most chargers, but does this make any difference when charging a Li-Ion? Faced with a choice of two USB>wall socket type chargers delivering the same 0.2A and 2A, does it matter which one I go for?

[HarryT]

It's the device which "decides" how much current to draw; the amperage rating of the charger merely designates how much current it's capable of supplying. I would imagine that the vast majority of devices would require more than 0.2A; a standard USB port can supply up to 500mA @5V.

[kennyc]

Any charger needs to limit the charging current in order not to damage the battery. Generally the higher current charger will recharge the battery faster, but might also reduce its life/number of charge cycles.

Here some websites:

http://batteryuniversity.com/learn/a...ion_batteries/

Says:

Increasing the charge current does not hasten the full-charge state by much. Although the battery reaches the voltage peak quicker with a fast charge, the saturation charge will take longer accordingly. The amount of charge current applied simply alters the time required for each stage; Stage 1 will be shorter but the saturation Stage 2 will take longer. A high current charge will, however, quickly fill the battery to about 70 percent.

http://batteryuniversity.com/learn/a...ased_batteries

wiki: http://en.wikipedia.org/wiki/Lithium-ion_battery

[tompe]

Originally Posted by HarryT

It's the device which "decides" how much current to draw; the amperage rating of the charger merely designates how much current it's capable of supplying. I would imagine that the vast majority of devices would require more than 0.2A; a standard USB port can supply up to 500mA @5V.

Yes, but without negotiation and USB port will give 100 mA.

But see the question instead as: Is it better to trickle charge or better to use a proper charger that charge fast?

I would think that it does not matter. What matter is that you do not discharge the batteries fully.

[HarryT]

Originally Posted by tompe

Yes, but without negotiation and USB port will give 100 mA.

That's true for a PC USB port; I'm not sure it's true for a "dumb" wall charger, is it?

[MissKittiness]

It doesn't matter whether your charging method provides 0.2A or 2A (as long as it isn't an extreme such as 210V into a 110V device), what matters is the EEPROM of the device which counts how many charges the battery wants to hold. You charging the battery at 95%, 50% or 3% makes no difference to the battery, however the EEPROM counts even a charge of 5% as a "charge cycle". So, the only real "best practice" is to wait until the device is lower than 10% before you charge it, and if you plug it in to the computer to load more books, you may as well leave it plugged in and deal with the "trickle charge" until it's full, because once it's plugged in, the EEPROM will count the charge.

[HarryT]

Originally Posted by MissKittiness

It doesn't matter whether your charging method provides 0.2A or 2A (as long as it isn't an extreme such as 210V into a 110V device), what matters is the EEPROM of the device which counts how many charges the battery wants to hold. You charging the battery at 95%, 50% or 3% makes no difference to the battery, however the EEPROM counts even a charge of 5% as a "charge cycle". So, the only real "best practice" is to wait until the device is lower than 10% before you charge it, and if you plug it in to the computer to load more books, you may as well leave it plugged in and deal with the "trickle charge" until it's full, because once it's plugged in, the EEPROM will count the charge.

This is not true with lithium batteries. They will NOT count a partial charge as a complete "charge cycle".

[MissKittiness]

Originally Posted by HarryT

This is not true with lithium batteries. They will NOT count a partial charge as a complete "charge cycle".

I didn't say that the battery counts it. I said that the EEPROM counts it.

[tompe]

Originally Posted by HarryT

This is not true with lithium batteries. They will NOT count a partial charge as a complete "charge cycle".

Yes, see e.g. http://arstechnica.com/gadgets/2011/...-ion-battery/:

One of the worst things you can do to a Li-ion battery is to run it out completely all the time. Full discharges put a lot of strain on the battery, and it's much better practice to do shallow discharges to no lower than 20 percent. In a way, this is like people running for exercise— running a few miles a day is fine, but running a marathon every day is generally not sustainable. If your Li-ion powered device is running out of juice on a daily basis, you're decreasing its overall useful lifespan, and should probably work some charging stations into your day or change your devices' settings so that it's not churning through its battery so quickly.

There used to be certain types of batteries whose "memory" of their total charge capacity seemed to get confused by shallow discharges. This is not, and never was, the case with Li-ion batteries. However, if you are using something like a notebook computer that gives you time estimates of how much longer the battery will last, this clock can be confused by shallow charging intervals. Most manufacturers recommend that you do a full discharge of the battery about once a month to help your device calibrate the time gauge.

One common misconception is that Li-ion batteries will only count charge cycles if the battery is drained completely in one session; another is that the battery counts one charge cycle for every instance the device is unplugged and plugged in again. Neither of these is true—Li-ion batteries actually count charge cycles based on a 100 percent discharge even when it's summed over multiple sessions. For example, if you discharge a battery to 50 percent one day, charge it back to 100 percent, then discharge it 50 percent again the next day, that is counted as one "cycle" of the battery. So shallow discharges, in all these regards, are ideal for a Li-ion battery.

[Belfaborac]

I should have been more careful with how I phrased things I guess. This is basically what I was after:

Originally Posted by tompe

But see the question instead as: Is it better to trickle charge or better to use a proper charger that charge fast?

By "better" think "making the battery last longer before requiring charging again". Never mind cycle count and whatnot, I'm only thinking of whether a single charge can be made to last longer by trickle charging rather than blast charging.

Originally Posted by HarryT

It's the device which "decides" how much current to draw; the amperage rating of the charger merely designates how much current it's capable of supplying. I would imagine that the vast majority of devices would require more than 0.2A; a standard USB port can supply up to 500mA @5V.

While that's true as such, it certainly takes a lot less time to charge from a 1A USB wall charger than from a USB port on my PC. Hence it (the PRS-950) must be able to utilise more than 0.5A if more is available.

[HarryT]

Originally Posted by Belfaborac

By "better" think "making the battery last longer before requiring charging again". Never mind cycle count and whatnot, I'm only thinking of whether a single charge can be made to last longer by trickle charging rather than blast charging.

No. Not with a lithium battery.

[elcreative]

Try googling on something like Li-ion manufacturer's specifications and you'll find at least one major set of information from the people who make the batteries as well as some stuff from a couple of physicists who did orignal research and design on Li-ion and Li-polymer batteries... personally I trust that info a lot more than that scattered from assorted articles scattered elsewhere...
Essentially a properly designed USB port will give 500mA which is available for a device plugged into it, this can be used in its entirety or at lower currents depending on the demand from the device... and the device/battery can also use much more for more rapid charging with the charge rate being reduced (to a near trickle level) once 80% full charge reached... Also the cycle percentage approach is the correct interpretation of charging cycles i.e. two 50% charges = a 30% plus a 70% = a 100% charge = one cycle... If you think differently, fine but I'm the one with Li-ion and Li-polymer devices that still charge fine after five years of regular usage using info from the manufacturers...

[carpetmojo]

All I know is the battery performances on my 350/650/T1 improved incrementally using a "never (well, very rarely) switch off, always charge on 1-2 at most, leave overnight, and use a wall-charger" regime.

Obviously it doesn't keep on getting better, but they all last a month, sometimes more, and much more on the 650 that doesn't get used as much as the other two.

I suppose the battery just gets trained ?

And none have died yet ....fingies x'd.

[elcreative]

Originally Posted by carpetmojo

All I know is the battery performances on my 350/650/T1 improved incrementally using a "never (well, very rarely) switch off, always charge on 1-2 at most, leave overnight, and use a wall-charger" regime.

Obviously it doesn't keep on getting better, but they all last a month, sometimes more, and much more on the 650 that doesn't get used as much as the other two.

I suppose the battery just gets trained ?

And none have died yet ....fingies x'd.

No training... that's following manufacturer guidelines which are actually right...

[MissKittiness]

Either way, you'll want to check to make sure there's no EEPROM controlling your battery's usability before you start experimenting.