Jon NCal said:
For an accurate comparison the internal cells need to start at the same temperature, suggest standard 68 deg. F.(20d C) Otherwise difficult to say how much variation is due to temperature.
I'm testing the pack as I'd use it on the bike. Each battery was tested the same way on my workbench and self-warmed during discharge to about 30C at the end of the test. Ambient temperature was usually around 15C year to year. Although I didn't specifically control temperature, I tested under similar conditions each year.
My notes for 2016 have ambient temperature of 15C for S1, 14C for S2, 14C for S3, 15C for S4, 17C for S5 and 18C for S6. In 2015 ambient temperature was slightly lower, 11C for S1, 11C for S2, 13C for S3, 11C for S4, 14C for S5 and 13C for S6.
The tests ran over a number of days. So, it's possible that S5 and S6 benefitted from being tested at slightly warmer ambient temperature in both 2015 and 2016.
I purposely do this test in winter when ambient temperature is lower because I want a pessimistic estimate. That way I can expect to get similar capacity in summer when temperatures are warmer and the pack a little older.
One thing I did notice was that on most of the packs cell 7 (the most positive) was consistently higher voltage than the other cells, and that cell 6 was consistently lower.