BQ77PL900: BQ77PL900 Cell Balancing

Hi Anton,

This is a very old device and there are only a few users. Normally I would suggest the BQ76930 or the latest battery monitor, the BQ76942 for a 6S battery.

Something to try: The input filter capacitor is double the value used on the EVM. It is possible the longer filter time constant is causing the voltages to measure incorrectly during balancing which could cause the wrong cells to balance. 

Best regards,

Matt

Hi, Matt
I tried to change capacitors value, and I also tried to use 510 Ohm resistors, but that didn't help
The balancing still strange

Hi Anton,

I assume you are using the cell balancing application report examples for this device?

Matt

No, I am not using the cell balancing application report examples for this device.
We work with LiPo batteries in our project, and we designed our own BMS based on BQ77PL900 and BQ34100 fas gauge. 
But the problem is that it doesn't work correctly as you can see. 
I actually don't understand why bq77pl900 balances each 4th cell and why it try to balance the least charged cell from time to time (the first cell). It seems, that bq77pl900 measure wrong cells voltages.

Hi Anton,

For the cell voltage measurements before balancing that you shared, how are these measured? By the device are with a meter? Cells above the OV - hysteresis should balance and the device should balance every 3rd cell. 

Regards,

Matt

We use stand-alone mode, where the BQ77PL900 IC measures cells voltage itself. As I understood from datasheet, the IC has to balance the cell, which is OV, to the voltage V = Vov - Vovh (Vovh - hysteresis voltage, which is 0,1V in our application).

So when I have one OV cell (for example, 6th cell) - the IC balance this cell and 3rd cell at the same time. But the voltage of the 3rd cell is less then Vov-Vovh. And the second problem is, that balancing didn't stop after 6th cell voltage became less then Vov-Vovh and this actually supprise me a lot.

How are you getting the voltage measurements you shared? Is this reported by the device or measured externally with an instrument?

I measure it with voltmeter

Hi Anton, I am not sure what could cause this behavior. Have you seen the same results on more than one IC?

Yep, I tested two PCBs and both of them were balancing cells in a wrong way.

Do you have a log of the voltage measurements from the device before and after balancing? It would be good to see what the device is measuring as compared to the voltmeter. Is the voltage measurement calibrated for the cell voltages?

Actually, we didn't programm host mode for bq77pl900. So we didn't implement cell voltage measurement from the device as we wanted to use less software for our pcb. So we could only measer the voltage across the battery with voltmeter.

Okay, I understand. The device will need to make decisions based on its measurements though so if the voltage measurements are off for some reason, it might be affecting the balancing.

It doesn't explain why bq77 try to balance every 4th cell (1 and 4, 2 and 5, 3 and 6) at the same time. This is really strange, because it seems to be some kind of internal balancing algorithm of bq77 IC. 

Hi Anton, The part should balance every 3rd cell (which I think is what is happening based on the cells you are listing).