Originally Posted by mrforsyth
Only if the cells in the pack have the exact same capacity, which is increasingly rare these days. If one cell has lower capacity than the rest and the pack is balanced during a discharged state, this one cell will then need to be balanced when the pack is charged or it will be at risk of overcharging. If balancing during the discharged state is avoided altogether, no balancing will be needed when the pack is charged.
It's for this reason that many of the higher quality chargers with carefully considered charge algorithms do not engage balance circuitry until the charger transitions to CV stage.
Thank you, but I am still missing the point. Storage and charge is relative to voltage not capacity as we know. If you want to store at 3.8v+- then you want all cells at 3.8v, regardless of slight cell capacity difference. So you want at least some form of balance on the storage cycle. At least you want an error on major swings to flag a bad cell or bad connector. If it a cell, you want to really check that pack anyway and possibly toss it. That is a good thing and what is does (actually you can run storage without the balance leads it stops on pack volts if that is what you want). Same with charging. You want all cells to charge to 4.2v+-. All cells will charge slightly different regardless. If it takes another minute or so to match, that is the price we pay to balance. If you just want to charge with no balance, you have that option. From my chair, it works pretty much the way you need in most cases and might error on the side of caution. The actual algo would be nice if we have the firmware programmer out there :-)