BQ24296M: charger selection and question for parallel cell charging

Hi Ricardo,

when Vin=5V and #cell=2s in series, bq25890/2 and bq24296M doesn't support 2s so we should select different IC.
the battery voltage range is 5.4~8.4V(2.7x2~4.2*2) so I think we should select one boost charger and one buck regulator.
Is my understanding correct?

In addition, in parallel condition, battery is removable independently so we need 2 port independence charger IC because we see short event when 3.7V(50%) battery and 4.2V(100%) battery is existing in the system.
do we have chrger IC supporing 2port independence charging?
if you know any way to avoide this short event with bq24296M or bq25890/2, could you let me know?

regards,
Kai

Kai,

Your understanding about 2S is correct. You would need to use BQ25887 and buck regulator to step down the output voltage.  We do not have any two port chargers in our portfolio. I could try to come up with a way to avoid the short circuit event, but unfortunately I do not have access to the lab this week. On both BQ25890/2 and BQ24296M the charge current is reduced down to 100 mA during a short circuit event until VBAT rises above the VSHORT threshold.

Thanks,

Ricardo

Hi Ricardo,

Could you explain following comment more detail?
"I could try to come up with a way to avoid the short circuit event, but unfortunately I do not have access to the lab this week.
On both BQ25890/2 and BQ24296M the charge current is reduced down to 100 mA during a short circuit event until VBAT rises above the VSHORT threshold."

Here is an example of removable battery configuration that I mentioned before.

Even if we reduce the charge current down to 100mA during battery-short, we cannot control battery-short event.

I come up with following circuit.
When VBAT1 < VBAT2, it looks no problem but when VBAT1 > VBAT2, V_SYS2 sees VBAT1 so I don't know this works well or not.

Regards,
kai

Kai,

In order to safely recharge a deeply depleted battery we autonomously reduce charge current to 100 mA when the battery voltage is below the VSHORT threshold. I thought it might help the parallel cells self balance if there was a reduced charge current, but you say that you are still seeing the short circuit event so this does not help you.

I do like your approach of adding a second charger and using a diode on the VSYS pin. I would recommend using a shottcky diode adding another schottky diode shown below. My concern is the power dissipation through might through the diode would be quite high. Would the load be 5A continuous or pulsed? Another solution would be to use two power FETs instead of diodes, but more hardware would be needed to drive the FETs. Would the customer be interested in this?

Thanks,

Ricardo

Hi Ricardo,

I have additional question about 2s series.

I realize BQ25887 doesn't have power path management and when it works as cell balancing, the charging current is limited by 400mA.

Customer requreiment is here

- power path management

- cell balancing with 2A charging

Do we have charger IC supporting their requirement?

If there is no single chip solution, are there any good combination to support it with multiple ICs?

Regards,

Kai

Kai,

For your customer, I think a two chip solution is the way to go. My only concern is your customer's system load might be too high for the diode "OR" be discussed to join the two SYS pins. Let me add in someone for power switches to see if they have any advice.

Thanks,

Ricardo

Hi Ricardo and Rakesh,

Regarding 2s series, here is brief block diagram of system requirement.

I find following apps note and if we use external MOSFET for Fast cell balancing, it seems to achieve 2A cell balancing.
www.ti.com/.../slua420a.pdf

however when we use boost charger like bq25887, it doesn't support buck. e.g 5Vin to 3.5V/2A cell balancing is not possible w/ bq25887.
I couldn't find any good solution for it but if you have any good proporsal, please let me know.

Regards,
Kai