The later CV stage of charging can take quite a while when fast charging due to the early switchover from the CC stage (due to the voltage rise from the cell’s internal resistance). The charger thinks the cells have reached a higher voltage than they really have.
Many BMS’ can, and do, balance when not charging as long as the cell voltages are high enough and far enough apart to warrant balancing.
These two things result in much more time that can be spent balancing.
We can’t think of the effects of cell damage that way. There’s no “official” voltage for the cells to have or drift from. The cell voltage is defined by the charger voltage and BMS during charging.
Three things can happen when a cell is damaged during discharge. It can develop a slightly higher internal resistance, it can lose capacity, or both of those things can happen.
The slightly increased internal resistance will force the cell voltage up some tiny amount during charge, perhaps 20mV if the cell is badly damaged by a discharge. This can force the charger to switch over to the CV stage a tiny bit earlier, actually allowing a tiny bit more time for balancing.
The increased internal resistance does mean balancing is needed but it’s such a tiny amount of charge to remove when you want to drop a cell by 20mV at full charge. At 3.6V it takes the removal of a much larger amount of charge but not at 4.2V.
A loss of capacity means the cell will charge back up faster than it did the previous cycle. But the loss of capacity due to one abusive discharge is going to be quite small. Moderate abuse of 30Q’s in my testing resulted in about a 2mAh loss per cycle for the first 50 cycles or so and about 1mAh per cycle after that. The small loss of capacity for even worse abuse can easily be compensated for by the BMS, even with the 8mAh limit you set in your example for the shortened charge cycle.
A truly abusive discharge cycle that results in a lot more capacity loss than that has probably damaged a lot of cells and the rider should probably consider replacing the pack soon.
There’s a reason why the BMS controller chips from the big companies like Texas Instruments, Analog Devices, etc., only have a relatively low spec for on-board balancing current. It is definitely due to the heat it creates but it’s also because more just isn’t needed.
We can set rarely seen conditions like ultra-fast charging with crappy cells and then, of course, more balancing will be needed. But even then I can set a BMS to balance both on and off charging and I can fully balance the pack every cycle. Once crappy cells are balanced they aren’t getting crappier really quickly. We actually don’t even need a lot of balancing current to keep up with the fading performance of a crappy pack until cells start dying.
If a cell starts failing, and rapidly changing its performance, then certainly the BMS might not be able to keep up with rebalancing the pack to compensate. Thatls fine IMHO because that cell should be replaced. It’s not worth trying to spend more and more time crippling the performance of the entire pack just to bring all the other cells down to the performance level of the failing cell.
There just isn’t a big change in capacity or internal resistance in one cycle. So high balancing current wouldn’t be needed.