Wow, I’m surprised by the amount of wear some of you have …
I still have 100% max capacity set because I use it throughout the day without the plug, and have been using my Framework since Batch 4 pretty much non-stop, and only charge it at night while I’m home. At work it is never plugged in, and generally is at 30%-40% when I get home. Using it 5-7 hours per day since October (6 months now), I have only 3.7% wear.
When was the last time the system went through a full discharge AND charge cycle?
After doing that, my tlp-stat -b is showing 97.4% capacity (or 2.6% wear). This is over about 5 months. Intuitively, this seems about par for the course (and certainly better than my previous laptops — with some of them, I had essentially killed the battery within a year because I wasn’t very conscious about unplugging from the charger, setting battery charge limits, etc.).
[edit] Never mind, I guess now it’s showing 96.5% capacity 
Either way, well within what I consider acceptable.
I do hibernate all the time, actually. I rarely shutdown or reboot my laptop, except for when required by updates. I also have a cap of 80% for charging, and I rarely let the charge get below 30-40% (basically, I reduce the number of full cycles the battery goes through).
Aha. I was wondering how to get this in Linux. Thanks @Chiraag_Nataraj, tlp-stat -b is reporting 98.5% capacity. This is a batch 4 delivered in October, so 6 months old.
Lately I shut it down, but for the first few months I kept it on the charger. I also set a charge limit of 80% when the feature was enabled (4 months ago?).
Seems pretty good overall?
I don’t know mate All I know is that I’ve gotten decent battery wear on this thing, far better than in the past (with my other laptops), and part of that is me being a lot more conscious of how I (ab)use the battery and not letting it drain fully down.
Five nights a week. I always plug it in at home, and my cap is still 100%.
Edit: OH nvm, I see what you mean, I haven’t let the battery go below 10-15% or so. Not sure how much that changes the max capacity reading though. Going down to 30% most days, maybe once it was at 5%? I don’t care to let it run that low though.
Edit 2: I also never reboot, and only hibernate. Not sure how that affects the reading by HWinfo either.
Keeping the battery in a mid-state of charge definitely does improve battery wear characteristics. See the picture below and compare the red line (100% to 40% SOC) to the green one (85% to 25% SOC). Both are 60% DOD, but the one that keeps the SOC in the middle has slower degradation (with lower DODs having lower wear rates). That’s why the lower battery limit is good for slowing wear (though evidently it didn’t help me).
(image source)
Also thanks everyone for posting. It does appear that my battery degradation is abnormally high based on your inputs. I’m working with support to try to find out what’s going on.
@OxyMagnesium I’ve also had my laptop for about the same time (Batch 1). Both upower -d on Linux and HWMonitor and powercfg /batteryreport on Windows shows my battery at 15% wear.
Odd (or perhaps confirming) that we have the same exact wear percentage, and I also think this is pretty high degradation for such a short period of time.
Curious to know your findings with support and let me know if I can help, run any tests, etc. I’d also like to get to the bottom of this.
Edit: I’ve kept my laptop plugged in most of the time. It has been on basically 24/7 apart from putting it to sleep/hibernate randomly, and I’ve had the battery charge limited to 60% since around whenever the relevant BIOS released. So there shouldn’t be a high number of battery cycles.
@Michael_Wu Very interesting. I agree that it’s quite strange that we both have the same battery wear considering that our usage has been very different. I’ve run mine unplugged most of the time, and I do fully shut it down every night so it can’t be either of those things. I will definitely let you know what support says regarding this.
@OxyMagnesium Huh, very interesting indeed. Our basically opposite usage leads me to believe that the amount the battery cells degrade just sitting over time is significantly higher than the impact of “normal” usage.
Here’s the serial number or whatever by the QR code on the battery, maybe needed to pinpoint a possible bad batch:
FRANBBATA
1121105BW
I could be wrong and this is just a theory at this point. Looking forward to it, thanks!
I’ve had a little concern with the battery wear level accuracy. I use HWiNFO64 to measure my laptop and the wear level varies every time I use it. Today it’s at 0.0%, the other day it was at 1.7%, on another day it was slightly higher. I have the charge cap set to 80% and don’t really run on battery very often.
Seems wonky.
Think it was in another thread that someone mentioned the battery is rarely in-stock in the marketplace, if ever. I wonder if this is partially due to Framework might be aware of the [questionable] battery quality issue (unconfirmed), and is trying to source from a different battery manufacture…without doing a recall.
Mind you, the 1000 cycles bar is typically under some degree of ideal conditions. Various temperature, discharge / drain rate, usage pattern will offset that.
I have seen better batteries / cells…ThinkPads, e.g. Sanyo & Sony (made in China). (i.e. it’s not a regional thing, but a battery manufacturer and possibly cell binning / matching matter?)
Batch 3 (Since Sept 29 2021) here and mine is showing 86.2% capacity (13.8% wear). I’ve been using it every single day on battery since I have college and work and the battery usually lasts me about 4-5 hours from full charge. I never capped it between 20% and 80%. 13.8% does seem a bit high on wear imo, not sure if it’s me or if it’s the battery.
A battery is a battery, subject to different applications and environmental conditions…but here’s a test:
Tesla battery researcher shows new test results pointing to batteries lasting over 2 million miles - Electrek
Battery Power Online | Accelerated Life Testing of Lithium-ion Batteries
I would like to see the following features in the BIOS:
@anon81945988 Yeah the wear reporting is definitely a bit suspect; I’ve also seen it fluctuate by as much as 2-3% over the course of a few days. I think its calculations depend on a lot on the depth and speed of discharge, so it makes sense that it isn’t super consistent. So far it hasn’t varied a lot more than that though, so it’s probably at least in the ballpark.
@Michael_Wu You could try removing your charge limit and continue using the laptop normally for a few days, maybe even let it discharge very low. I did this at the suggestion of support and it does seem to have some effect on the reported wear, and though it didn’t change much for me perhaps it will do more for you since your limit is lower.
@OxyMagnesium thanks for getting back to me!
So yeah AFAIK/IME if there’s a charge threshold, over time, the battery wear percentage will drift away and won’t be as accurate as after a full 0 to 100 charge cycle reading.
I think (unsure) I did at least one full charge cycle between my last post and now.
Currenty it’s showing 87.43%, more than the prior 85%.
If accurate, I think 87% battery wear in 8-ish months (while being plugged in almost 24/7) isn’t too great.
I’ll perform a few full 0-100 discharge/charge cycles and report back, hopefully it goes up even more!
Edit: typo
Um, after 17 days, wear increased to 9.350% now…
Edit: I was using 80% charge max, for about a week, yet still 
Over 2.5 months from new and running Ubuntu 21.10 and now 22.04 and only charging to 80% the battery is showing 2.5% wear. Used 80% of time plugged in and over night the laptop is shutdown and unplugged, Running on battery I am getting around 6hr use on “Power Saver”
Monitoring using Linux “Power Statistics”
Shouldn’t have any impact at all since laptop is running off AC power, but battery wear readouts are weird.
