BQ77PL900: Balancing questions

Hi Jake,
1. Yes, when used in standalone mode the balancing operates at the overvoltage threshold and until the cell drops below the OV hysteresis threshold. It does not matter if the cell is charging or discharging. Normally charging is required to raise the voltage to the OV threshold, and normally a moderate discharge will pull the cell voltage down and cause an OV recovery. If the OVH is set to zero or the charge current is high when the cell goes into OV, when charge current stops the cell voltage may immediately drop below the OVH threshold and there may be no balancing.

2. No voltage difference or imbalance threshold. The part will balance at the OV level in standalone, it has a tolerance shown in the data sheet. When using host mode the balance decision is made by the host, calibration of the cell voltages will have a large effect on the balance.

3. It looks like the readings are at the cell and perhaps at the IC. It is unusual for the top cell to have such a large difference. I would look for leakage into the VC1 net, either some path to GND or unexpected current into the IC. If current were in VC1 and out VC2 the VC2 & cell 9 voltage should be larger. Inspecting the voltage across the input filter resistors can be helpful in these situations, it is normally very small with peaks when the part samples voltage.