Why do Aeotech Multisensor 6 devices go from 100% batteries to dead in weird ways? EEs?

I’ve had 9 of these for over a year. I have never once seen some sort of average decline from day 1 to day 100, 100% to 0%. (Their batteries last 96 days for me, on average.)

I’ve always seen them do weird things. They mostly stay at 100% a long time, then drop down, probably to 50%. And when they fall below 50%, they die within days. A kind-of typical pattern is to be at 100% until day 75, then drift to 50% for a week or two. Then a day of 40%, then dead.

But I can’t describe all the oddities I’ve seen; 3 months X 9 = 36 graphs. One sat at 1% for weeks before dying.

I have logs if there are specific questions, but it’s a chore to summarize.

I can’t rightly explain this except to think that there is some super simple testing of battery strength.

Hoping some microelectronics Electrical Engineer can chime in?

I have an aeon multisensor 5 that does the same thing… and I have yet to find out why.

My untested theory so far is that the configruation of the switch changed for some reason (e.g. shorter reporting interval).
Tool to use for this: