Before charging the pack, it is important to balance the batteries. Each cell (I’m using the term cell for what is actually, in the Leaf modules, a pair of cells in parallel) should be equal in voltage to the nearest hundredth (0.01V). An ordinary voltmeter can be used. Mine is a $10 multimeter and it gives very consistent readings.
Of note, my multimeter will either display VDC in hundredths or in thousandths. In either one scale or the other, it is perfectly consistent. But there is a fixed value difference between the reading it will give in hundredths, compared to the reading in thousandths. So in my case, it is best to compare cells using one scale or the other. It doesn’t matter which one; the absolute value is not as critical as getting the cells the same.
To tweak the voltage of a cell up, I got a small single-cell charger. It only charges at 2A, but that is enough for small adjustments. It provides a lithium profile (constant current, then constant voltage) for 4.2V max charge, which is the max for this battery chemistry. So, it is safe even if I accidentally left it charging the battery overnight. But I only used it for initial balancing. So far, I have not had to re-balance the cells. If I did, this is what I would use to bump up a low cell.
To tweak the voltage down, I made a small load with two flashlight bulbs in parallel. They are bulbs you can buy at Radio Shack for 3.6V (3 cell) flashlights. It is very crude and I intended to make a nicer device later, but since I have only used it once for initial balancing, I never got around to making it pretty.
I balanced the cells at 3.5V, which is closer to the bottom end than the top (the operating range is 3.0-4.2V). After fully charging the pack, they are always within 0.02V of one another. After fully discharging the pack (riding until pack voltage is 22 x 3.0V = 66V), which I do very seldom and only to test the max range, the cells are within 0.03V of each other. The reason for the difference is that the ‘cliff’ on the low end is quite steep at 3.0V.
When I charge the pack again, the cells return to within 0.01V of each other in the mid-range and 0.02V of each other at the upper end.
I charge them to 4.14V (91.188V ÷ 22 cells), and when charging is complete, they settle to 4.12 (90.7V pack), so I have a little safety margin at the top.
Note: Conveniently, the AXE controller interprets the voltage as being within it’s 90V upper operating limit, so I can turn on the controller immediately after charging the pack.
Next: Bench Test