Lithium Ion battery charger

Thread Starter

Experimentonomen

Joined Feb 16, 2011
331
I'm working on building a charger for a 22.2V pack i made with six 18650 cells from a laptop battery back wired in series.

Im considering active balancing by using the shared transformer method, see fig 7 in this pdf: http://users.utcluj.ro/~atn/papers/ATN_2_2010_1.pdf

I'm thinking a 30 turn primary and six 5 turn secondaries as that would turn the full pack voltage into the correct voltages for each cell.

The part i don't quite understand in that pdf is that the secondaries are wired the wrong way, the diode is facing the negative pole rather than the positive pole.

Is this a drawing error or does that reverse current act like the passive shunt balaner by using a reverse current over the cell to divert the charger current to the other cells ?

And i imagine the box that sez "control" is more or less just a 50% duty cycle clock that gets an enable signal from the charger.
 

wayneh

Joined Sep 9, 2010
17,495
Is this a drawing error or does that reverse current act like the passive shunt balaner by using a reverse current over the cell to divert the charger current to the other cells ?
IMHO, it's not an error. The diode diverts current around the cell rather than thru it. But I don't really understand why this arrangement wouldn't drain the cells if no charge pulses are applied. You should write the author.
 

Thread Starter

Experimentonomen

Joined Feb 16, 2011
331
Ah i never realized that, with no switching signal on the mosfet, the diodes and secondaries actually impose a dead short onto the cells.

I should just be able to flip the diodes and doing the charge current limiting primary side and sensing the total pack voltage for the constant voltage regulator. It should provide a more balanced SOC than just a plain old series charger.

Since all six secondaries have the same number of turns(like a six strand litz wire) would produce exactly the same voltage, there should be no need to monitor the voltage of each cell.

Its just a matter of limiting the primary current so that each of the six secondaries only supply up to 1.1A of current to the respective cell.

This current limiter could be a current shunt in the switching inverter feeding back a current signal to the error amp that controls duty cycle and thus controlling the voltage of the six outputs, limiting the charge current.

Or simply a current transformer on each output, feeding back to the PWM controller, now i dont have six identical small ferrite cores though.

I'll have to build a prototype and test this out.
 

wayneh

Joined Sep 9, 2010
17,495
Ah i never realized that, with no switching signal on the mosfet, the diodes and secondaries actually impose a dead short onto the cells.
In Fig. 7 it looks that way, but that's not really a schematic. There may be more to the story, one of those "details left to the reader" sort of things.
 
Top