I have been running on battery far more, though with a lower 75% charge limit. Probably another ~100x 75 to ~25% cycles in the last two months.
After charging to 100% and letting it float down to 80%, then back to 100% 3x times today (to let the charge controller figure things out), I am getting 3350000/3572000 health (as reported by cat /sys/class/power_supply/BAT1/charge_* ), or 94% of rated capacity. This means around 4% capacity loss in 4 months since my last measuring and posting on this in April.
Others have said that the battery state of health calculations might be a bit inaccurate, and my suspicion is that this is true as well. Not sure how to improve the accuracy of the reporting, either installed in chassis or through bench testing, but it might come to that if we need better degradation data.
So that’s 50 cycles. I have only done 51cycles in six months with 7% loss of capacity.
Not sure that is going to work. To calibrate the battery it will have to be taken down as close to zero as possible and up to 100% for the calculation of what the battery can hold.
Over ≈7 months I have calibrated the battery reporting algorithm 3 times and each time the result shows lower wear. So much so that each successive time the wear been reported as dropping.
This would also make sense if the discharge to zero and recharge to 100% ‘stretches’ or conditions the batteries capacity.
The other note is that the battery cycle increases as expected.
I am at 98.2% after about 7 months. Running Ubuntu 22.04.1 spending a lot of time on power with a charging limit set to 80% and running on batt about once per week and getting around 6 to 7 hrs use from the battery.
I was in batch 3 and was using mine daily for a day short of an entire year now (Got it Sept. 29, 2021). I’m at 89% capacity so 11% wear.
I’ve been using it daily for college and work and charging to 100% and usually using it until it’s around 20-30% charge though some days I went as low as 5-10% due to not having access to a power outlet.
Batch 2 i7-1165, 97% battery health remaining. Semi-regular daily driver. Charge to 100% and drain from there. Variable drain lengths before re-charging depending on how long I know I’ll be from a wall socket.
Not accurate. At the end of the day, what you really / actually care about is how long you can use your laptop for, per charge. That’s what really matters. Wear level is just a rough guesstimation.
It’s like…mileage of a car…it’s not exact…and what really matters is you want to know if your car will crap out on you, and how much it’ll cost to fix the issue. Mileage is just a rough indicator of likelihood of potential issue.
Maximum charge at any given time (MC) To find this the battery requires a complete discharge and a full 100% charge
MC divide by DC will indicate the wear on the battery in terms of capacity.
Other wear is harder to assess
Heat damage by fast charging may be breaking down some insulation and that sort of wear is more difficult to assess. One way of measuring it would be to measure the battery charge low when not powered on.
This type of degradation id more often liked to the cycles that a battery goes through.
So we have wear as in capacity (CA) and wear as in cycles (CY)
I’ll let someone else find a formula for their interaction
Technically true because they are at least charged (once or more) at the factory.
However, a lot of the time, a [good] new battery would be able get a full charge capacity anywhere between 1-5% above designed capacity. My Framework laptop was able to get a full charge about 1% above the designed capacity for the first 5-8 cycles.
I joined this thread not so much to mention my battery wear but to comment on the fact that replacement batteries are still not available to buy 15 months after the first computers were shipped. That is incredibly poor logistics.