Li-ion charger

[zachtheterrible]

I've got a Toshiba PDR-BT9 battery that is 3.6V and 900mAh.

I looked into using the LT1734 IC to charge the battery, but the datasheet specified that it only comes in 4.1V and 4.2V versions.

I was wondering if anyone could offer an alternative to charge a 3.6V Li-ion?

 

[Odysseas]

Afaik Li-Ion batteries can ONLY be charged with the appropriate voltage with a tolerance of only one single percent. I'd say you've gotta go for a 317 and adjust it properly. But don't take my opinion too seriously

 

[justDIY]

Zach,

you need to look up the specs for your toshiba cell and find out which it is, a 4.1v or 4.2v recharge voltage ... the recharge voltage is always going to be higher than the discharge voltage.

when in doubt, choose the lower voltage, you may lose some battery capacity, but you won't have it go all 3-mile island on you either.

 

[hjames]

Early lithium-ion cells came in the 4.1V ratings. I'm pretty sure that every Lithium-ion/poly cell that's been made in the last few years has been 4.2V.

The oddball rating is where the voltage discharge plateau is - when the battery discharges, depending on the load, it'll quickly decay to somewhere about ~3.7V and then keep on slowly decaying down to below 3V. If you discharge the battery too far below 3V, it'll permanently damage the battery.

Obviously find a data sheet if possible, otherwise look for some battery that "looks" close - Panasonic makes some lithium cells and has some data sheets/app notes on them. Read up and ask questions - it's better than exploding/killing a cell..

 

[Tarsil]

I've got a Toshiba PDR-BT9 battery that is 3.6V and 900mAh.

I looked into using the LT1734 IC to charge the battery, but the datasheet specified that it only comes in 4.1V and 4.2V versions.

I was wondering if anyone could offer an alternative to charge a 3.6V Li-ion?

What's the voltage of the fuly charged batery?
It might be 4.1V or something like that.

 

[justDIY]

Hey Zach,

Check this little gem out: **broken link removed**

Oh, and this one is even cooler: **broken link removed**

 

[RadioRon]

Lithium-Ion cells should be charged like this. Assuming the battery is discharged, begin charging using a constant current source set to deliver the current specified by the battery manufacturer and set to a maximum voltage output of 4.2 Volts per cell. Typically a charge rate of 0.2C is acceptable but in some cases 0.5C is OK. As mentioned the maximum voltage of the constant current source must be set to the peak charge voltage of the cell, typically 4.2VDC. As the battery charges, its terminal voltage will rise until the source can no longer deliver the constant current and it will go into constant voltage mode. From this point on, the charge current will decrease slowly. Charging is completed when the charge current drops to a low value, typically about 10 mA. At this point the battery is fully charged.

 

[hjames]

...
Typically a charge rate of 0.2C is acceptable but in some cases 0.5C is OK.
...
Charging is completed when the charge current drops to a low value, typically about 10 mA. At this point the battery is fully charged.

RadioRon's correct except for one point - charging is usually terminated when the charging current reaches some fraction of the battery capacity, typically "C/10". For a 1AH battery, this is .1A, so forth. Trickle charging Lithium batteries usually leads to fire...

Charging currents are usually a bit higher than .5C. I think 1C is pretty standard - consider that most Lithium type batteries recharge in ~2 hours or so, the first hour gets it to ~80% - 90% capacity, and the remaining time inches it closer to 100%. If the battery is really low though, there is a preconditioning cycle that occurs though.

reference:
https://www.panasonic.com/industrial/battery/oem/chem/lithion/index.html

 

[zachtheterrible]

Thank you much for all the information, its been very helpful! I'll have to search the net for the datasheet to this battery. Those maxim IC'S look very apealing.

In regards to not letting the battery dip to a certain voltage (which will be specified in the datasheet), how do I do that?
I was thinking of putting a FET in series with the entire circuit and biasing it so that if the battery drops below a certain voltage, the FET won't let current through and ruin the battery. I was thinking of using the FET because of it's high input impedance, so it will use a negligible amount of current, unlike if I used a bipolar transistor.

 

[hjames]

A mosfet works - just keep in mind the parasitic diode. Battery disconnect circuits typically have two mosfets in back-to-back to make sure current doesn't flow back into the battery.

The "easy" solution is to just get a battery protection chip. Lots of companies have them - Texas Instruments is the main one that comes to mind, but there are other companies. The main benefit is that these are pretty self-contained solutions - they'll disconnect the battery when overcharging, under discharged, over current, over temperature, etc. They don't make the battery bullet proof, but you'll have to work a lot harder to mangle a protected pack. The main issue is that they'll add a ~.1Ohm to the circuit or so.

 

[zachtheterrible]

Thanks h james. I can't find any kind of datasheet for my battery, apparently toshiba stopped making them, and when emailed they said that they don't have access to that datasheet anymore.

I think Ill be alright in making a charger, seems extremely simple with those wonderful maxim chips.

Now I have a bit of a pickle before me. I'm wondering if the battery that I have might be a "smart" battery, one that might not need that protection chip. I got the battery from my local electronics store for 1$ so I have no problem in wrecking it.
I was thinking of discharging it until it will no longer put out any kind of voltage (it either killed itself or the protection circuit kicked in). Only problem is I have no way of recharging it to see if the battery is indeed smart (it didn't kill itself). Any ideas on how to find this out?

By the way, how does the protection chip monitor the temp of the battery? I see three connections to my battery. Does it connect to that?

 

[hjames]

heck, here I was thinking you had a bare cell. This thing most likely has a built in protection circuit - If it has a CE sticker, it's gone through at least some testing, if it has a UL sticker then it's gone through a heck of a lot of testing.

There's two possibilities, either the third pin is simply a thermistor (probably connects to ground), or it is a single wire interface to an on-battery protection/counter chip - google for "HDQ" or "SDQ". Short of opening up the battery, it would be hard to figure out. "Intelligent" batteries usually have the temperature sensing built onto the controller chip, and only accessible through a software protocol.

In any case, If you discharge the battery down to ~2.5V, the battery will either keep on dropping (i.e. it's not protected), or it will suddenly go open circuit. When you charge it back up, it should accept the charge current properly and reconnect itself. Dropping the voltage to 2.5V does damage the battery a little bit, so it might just be easier to (carefully) open up the battery...

 

[Nigel Goodwin]

heck, here I was thinking you had a bare cell. This thing most likely has a built in protection circuit - If it has a CE sticker, it's gone through at least some testing, if it has a UL sticker then it's gone through a heck of a lot of testing.

You've got a LOT of confidence in stickers!.

Chinese manufacturers will put any sticker you like on the equipment, just for the cost of the sticker - no testing at all done!.

Don't know about UL?, but CE is self certified - and there's no specific tests specified!.

 

[justDIY]

dell battery packs have every kind of certification known to man on them, yet some still went "kaboom". on my dell pack (with sony batts) I see CSA, UL, TUV, CE, (D), (FI), (N), (S)

 

[Nigel Goodwin]

dell battery packs have every kind of certification known to man on them, yet some still went "kaboom". on my dell pack (with sony batts) I see CSA, UL, TUV, CE, (D), (FI), (N), (S)

I heard the other day that the fault was in the plastic surrounding the batteries?, supposedly there were metal particles in the plastic making it conductive?.

 

[hjames]

You've got a LOT of confidence in stickers!.

Chinese manufacturers will put any sticker you like on the equipment, just for the cost of the sticker - no testing at all done!.

Don't know about UL?, but CE is self certified - and there's no specific tests specified!.

Well, CE is self certified, but the certification lab I went (to to do some testing on a LiPoly based device) did the obvious tests - overcharging, short, simulated component failure. An actual UL label on a battery actually means a lot of testing - a whole slew of tests including crush and puncture(!) testing.

Obviously a noname knockoff part will be inherently suspicious, but Toshiba isn't exactly a fly-by-night outlet.

As for the "slew" of battery failures out there, it's not exactly common - a handful of failures for the millions of Lithium cells out there? You're more likely to be hit by (vehicle of your choice) than have your laptop take out your family jewels in a freak accident - plus Dell isn't exactly synonymous with "robust hardware". In any case, this whole mess means that battery manufacturers now know what *not* to do, so hopefully this becomes even less of an issue than it is now.

 

[justDIY]

Zach,

Be sure to keep us updated regarding your exploits of those Maxim chips ... I'm brewing a project right now based on the max1811.

 

[audioguru]

Millions of compact fluorescent light bulbs were recalled and replaced by the importer here. Some of them shorted then leaked flaming plastic. The newspaper said that the Chinese company that made them stole the certification labels from another product.

 

[Nigel Goodwin]

Millions of compact fluorescent light bulbs were recalled and replaced by the importer here. Some of them shorted then leaked flaming plastic. The newspaper said that the Chinese company that made them stole the certification labels from another product.

Why steal them?, they just print their own.

 

[justDIY]

We all seem to have hijacked Zach's thread here ... it's no longer about lithium battery charging, the last page of posts has been about the questionable values of the certification process and counterfeit labels.