Protection circuit for Charger IC for Li-ion battery

[Kay]

Need to make a circuit to charge Li-ion battery, input is a DC adapter, ( Constant current and constant voltage adapter could be used, The input voltage can be up to 8V).

I am aware of the Charger ICs that support up to 8 or even 10V input but I would like to cut costs as much as possible so I am trying to use low cost charger ICs with 6.5V input like APL3202 or STC4054.

I made a little circuit to act as a pass element, if the input voltage be higher than 6.5V then the extra voltage would drop on the pass element. Built the circuit and works great in the lab, it does what I need.

The circuit is supposed to be used in the production units as well. I am not  professional in electronics so this circuit may have some inherent issues that I may not be aware of so any input/ feedback/ or possible issues on the circuit would be greatly appreciated. 

How the circuit works: The resister voltage divider( 6K,1K) sense the voltage on the charger ICs, if the voltage goes higher than a certain amount then the transistor turns on and then pulls the gate of the MOSFET and causes it to take the extra voltage on its Vds.

[digsys]

You need to get rid of ~10V-6V x 0.5A of heat, so why not use a 8ohm 5W-7W resistor in series,
with a 6V2 5W Zener across the device. The resistor and Zener will dissapate 2W of heat each max.
Whatever you use, you will need to dump that amount of heat at max charge anyway.
No need to  be simpler than that.

[nctnico]

All Li-ion battery packs I've come across have internal protection circuitry. AFAIK you can't buy bare Li-ion cells without signing a non-liability contract from the manufacturer.

[digsys]

All Li-ion battery packs I've come across have internal protection circuitry. AFAIK you can't buy bare Li-ion cells without signing a non-liability contract from the manufacturer

Hiyas,
This is another of those "grey" areas. There are rules governing disposal, use in public, transport on public carriers or anywhere they can present a
hazard. You can experiment with your own BM, and use it, but IF something goes wrong, you'll need to prove that your BM was up to spec.
We buy 100s of OEM LiION cells and make our own BMs. MOST of the manufacturers BM modules I've seen are pretty cr@ppy, and they're approved
for use .. mostly in EVs and that's serious. Basically, take EXTRA care when you design / test it and / or use it in public. Especially if it will be used in an
unsupervised situation ie alarm system backup. A (serious) question we always ask ourselves when doing an event is .. "what do we tell the coroner?"
There are a LOT of LiION incidents every year, remember all the Notebooks that were catching fire. Playing with LiIONs is great, but READ the rules.

[T4P]

BM = ?
There are lots of unprotected Li-Ion cells everywhere

[M. András]

battery managment "system"
you can even bare li-po cells from a few sites but you gonna pay a shitload of money to the secure containers they ship them. i dont even know why they would need an explosion and fire proof container for 60 bucks to ship a damn cell. in its original package i guess you cant short the terminals cos they are nowehere near eachother and plastic/paper does not conduct electricity. as for heat which curier company ships things next to a furnace?

[Kay]

Digsys, Thanks for the comments.
For your suggestion of using resistor, 8 ohm and a Zener:

Our device is supposed to work with any Nokia charger in the market,we don't ship the charger,  the charger come in Current Limit format, their voltage ranges from 5 to 7.5 volt or even higher and maximum current at about 800MA but generally they are at 300 to 500MA,  no standard there as far as I know considering all the aftermarket chargers so we set the maximum allowable voltage at 8V.

Now if somebody is using a charger at 5.5 volt then the charger can pump its current at about 5V or less, assuming 5V. Now when battery is reaching 4V-4.2V then we have less than 1V to play with, (0.1V goes for the drop at the charger IC as well), so maximum current is now becomes less than 1/8=125MA so there is an increase in charge time there.

so ideally it is better to take the extra input voltage when it actually exists so I made that circuit, but does the circuit look professional?

For the battery protection we use standard protection boards in the market and the Charger IC also has another level of protection  so 2 levels of protection there and we ship the battery in a separate box, battery is not shipped inside the device, rules are certainly very important for Li-ion, we even considered Li-fepo4 but difficult for the consumer to find a replacement.
 
 

[digsys]

.. when battery is reaching 4V-4.2V then we have less than 1V to play with, (0.1V goes for the drop at the charger IC as well), so maximum current is now becomes less than 1/8=125MA so there is an increase in charge time there.
so ideally it is better to take the extra input voltage when it actually exists so I made that circuit ..

Well, now you have a problem. There is a universal charge profile for CORRECT charging of LiION. You need to check the manufacturers papers for exact
voltages and currents, but basically it goes -
Max of 1C (come 0.5C - 2C) via constant current at a max voltage eg 4.5V 4.25V (corrected mistake), then WHEN the current drops C/10, the charger must drop to constant voltage
eg 4.20V max with current limit of C/20. Generally this changeover point happens ~3.8-3.9V , at the cell, so 125mA at 4.0V is most likely TOO high.
All depends on chemistry type. If you look at the charge / energy curve of LiIONs, there's bugger all capacity to be gained after 3.8-3.9V anyway, so
VERY few designers bother to gain that extra 2-5% and RISK a failure. You can PUSH it, but you WILL SEVERELY reduce the total # cycles.
In a race pack we DO push it past the limit, and if we only get 10 cycles, we're happy, especially if it helps win a race (or gain a few positions).

[amyk]

Max of 1C (come 0.5C - 2C) via constant current at a max voltage eg 4.5V, then WHEN the current drops C/10, the charger must drop to constant voltage

I've read that lion batteries should never have more than 4.25V applied to them...

[digsys]

I've read that lion batteries should never have more than 4.25V applied to them...

Someone stole the 2 .. bastids :-)