Given the variable wear figures I have, which seemed to correalte to the charging maximum I decised to let the laptop run down and switch off. I then plugged it in and left it all night to charge to 100% for some 8 hours.
I had originally set the level to 90% and as above and had 3.6%
After a couple of months I set it to 78% and the wear level went up to 7.8%
One day it went down from 7.6% to 6.7% overnight
Today at 100% overnight it now reads 5.8%
Will keep monitoring.
By the way I note the red light is flashing at 100%
Unplugging power and pluggin in again light goes to steady white.?
Will run plugged in at 100% for the day and see what readings are tonight.
Update: Evening
Ran down to 70% and set charge limit back to 78%
Update: Next evening
Wear down from 5.8% to 5.6%
UPDATE:
5 days later it is down to 4.9% ???
One thing that seems to have had an effect was allowing the laptop to run down and switch off and then charging to 100%. Will do the same on the 1st of next month too.
I received my Framework DIY laptop a couple of days ago and when I ran HWInfo64 I noticed that the battery had 2.2% wear leaving it at about 54WHrs rather than 55WHrs. I was just curious if anyone else noticed if their Framework laptops had wear already on the battery?
In win 11 using powercfg /batteryreport I can, by calculation, that the reported battery wear is 100 minus the reports Full Charge Capacity / Design Capacity, which align with those from HWiNFO64
Wear percentage
28th Feb
At start this was 3.4%
A week later after it had improved to 2.9% but I hadn’t noticed
After 3 months it increased to 7.4% 4.5% wear in 3 months
June 1st I run the battery dead and charged to 100%
After a few days it stabilised around 4.9%
so 3% wear in 3 months
June 11th:
Allowed to run down, switched itself off with capacity around 5.6%. Waited 20 min switched back on, ran for another 30 min and switched of with capacity at 2.8%.
Charged to 100% and run whilst plugged in for an hour: wear at 5.6%
12 hours later after no use down to 5.5%
Another 24 hours later after no use down to 5.4%
June 22nd 6 days later it is down to 4.8% So a loss of 1.4% wear from beginning or 1.9% from best resording
June 27th discharged battery
June 28th Charged battery to 100% At the end of charge the wear was 6.1% **
12h later** used whilst plugged in wear is 5.1% 4 months old
July 1st wear records as 5.5% July 6th wear records as 5.6% July 10th wear records as 5.3%
Given the initail wear of 3.4% this equates to only 1.9% in 124 days or about 5.6% in a year
Am going to keep a record of cycles now we have the 3.09 BIOS
Batch 8 just over 3 months use and running Ubuntu 22.04, power consumption management by “LTP 1.5”. charge limit set to 80% and “Power Statistics” is now showing around 2% wear, this value tends to range between 97.8% to 98.2% capacity not sure why maybe down to ambient/battery temperature.
The laptop will run for around 6hr on the battery.
I’m going to another today/tonight/tomorrow. I was going to wait for a month but as I see the changes I’m going to see what happens after a full cycle ten days later
I did a full charge and KDE Info center reports my battery as 46Wh as a full charge out of 55Wh. It shows a 86% battery capacity. Maybe I got a bad battery?
That wouldn’t do anything, though, since the underlying programs and systems are completely different (the shell is different from e.g. powercfg and the two aren’t related).
Ok, so running upower -i /org/freedesktop/UPower/devices/battery_BAT1 reads charge-cycles: 217. The Framework Battery page reads “This pack is designed to provide up to 80% of its original capacity at 1000 cycles”. My battery is at capacity: 85.3303% after less than a year.
With me being someone who reads into every line, between lines…etc… the 80% is best case scenario given the way it’s worded…and only by “design”. i.e. Actual production / implementation result may differ. And it doesn’t state the worst case, or nominal case expectation.
What’s missing from Framework publicly (to my knowledge) is what’s the battery replacement policy under warranty period?
The 80% is seen as the failure point for a battery, though if it can last 2 years to get to that then why not a few more years to get to 50% and for some 50% will be fine for some users.
Then there is the use, how much load do you have. After some 4.5 months I read 45 cycles and I even though that is a lot as I use it plugged in most of the time. However plugging in whilst using doesn’t really make that much of a difference, from previous experience too.
Although the main board can run on an external supply only when the battery is connected it will supply the power, so it is both supplying and receiving power when connected. This may keep the battery at a set level but doesn’t stop it’s use,hence my 45 cycles when if I were to measure the times is was only on battery and so getting ‘discharged’ I would have thought it may be around 10 OK less than 20.
Then there is the initial capacity. I used [powercfg /batteryreport] and noted it started off at 3.6% wear, it is now 6% so it’s really only 2.4% wear.
So if cycles is a representative of wear then the cycles on mine are between 24 and 60
How is the cycle counted? If it is a coulomb count then I see that it is just a math of max cycles ÷ wear, but is it? A lot more speculation can follow …
Anyway at two months my wear was 7.4% which according to my speculative math would have given me 74 cycles. At that level of use 10months would have been 740 cycles.
I decided to try and calibrate the battery.
I let the laptop run the battery down until it powered off then charged to 100% (I had it set to a max of 90% in the BIOS)
The result was that the wear was then recorded as 6% and after a few days went down below 5%
I have done this three times now.
I run the battery down, turned off just below “5%” wait ten min and restart getting another half an hour and it turned off again this time at “2%”
Clearly the percentage is a calculation and is not the same as the voltage which is the value required.
So now at 4.5 months I have 6% use.
Consider your use
I don’t play games and half my work is in the evening in a very low lit environment. Even in the day the general demand is only about 12 - 13 watts. So you may like to check how much power you are using. The more you use the more that battery is used even if plugged in.
Well to counter my speculative math the count is not a function of wear according to reading I am looking at now.
Right now I have a 5.3% wear and still 45 cycles.
So maybe the cycles is not that important.
If I were to take the 5.3% and take away the initial 3.6% I have only worn the battery 1.7% in 4.5 months. So 17 cycles by my short sightedness, but the cycles hasn’t changed ???
So on a good day like today it seems as though I may only get 5% wear a year, though no doubt the count will sober me up.
Try calibrating the ‘battery’ occasionally maybe.
So your loss of 15% in a year reeks of a lot of use and with my dumb math would equate to 150 cycles (of the 1000)
So your 15 ÷ 217 and my 6 ÷ 45 are a bit out giving me half the count per percentage.
Apologies for my layout. I just write as I’m thinking, which is messy etc.
So in 24 hours the wear has dropped ??
A more clear layout of my findings are in post 66 above
