Are there any plans to improve on the current battery charge limiting behaviour?
For example, I noticed that when I limit the charge to a particular percentage, the laptop will discharge the battery until it goes back below the chosen charge threshold.
Instead of doing this, my previous laptop would just not draw power from the battery at all and just run off the mains! The battery would then just discharge at the rate it would if the laptop was switched off (which is very slowly!). This meant that charge cycles happened much more slowly on that laptop - whereas with this one I’m already on 16 after having the laptop only three months, and my capacity has gone down by 4%! For comparison, my old laptop is still at 95% of its original capacity after 4 years (albeit, most of those years were with the laptop being plugged in)!
I’m aware I can always switch off battery usage in my BIOS, but ideally we would be able to do this without rebooting or manual intervention when trying to limit battery charging.
If there are no plans for this, is it something the framework team would consider implementing? I think it would really help to improve the longevity of our batteries.
I have 61 after 12 months so that is similar.
Framework state the battery can take 1000 in 2 years and still be 80%
At my rate I may only use 100 in 2 years
Regarding the wear.
You may have to recalibrate the measuring programme by discharging completely and then do a 100% charge a few times.
Do you have a record of the intial wear, batteries are often shown as having wear on day 1.
Mine was 3.4% on arrival and went up to 7% in a few months before I intervened; now it is down to 3.7%
Please see the graph and read more in this topic.
Further look at my details. Wear is less in the morning once the battery has rested. Wear is less, usually, every time I do a full discharge and full charge. This won’t happen for ever and the measuring can’t be trusted as accurate, but it does given a relative indication.
That seems exactly as I would expect.
Each month I charge to 100% from zero and then set the limit to 78%, so the battery will run down to 78% before charging.
I don’t want it to stay at 100% so it would be a problem if it didn’t discharge.
I understand you rather the mains power be used instead of the battery (when charge is greater than set limit) but unless you are changing the charge limit you would not be in a situation where the battery % is greater than limit % right?
Perhaps lodging your desires with support might be more helpful to get an answer as this is a user forum with limited Framework staff engagement at least nowadays.
I’d second some support here… I have no idea what the tolerance is when a charge limit is set, but my OS is getting constant ACPI notifications of battery state changes (at least once per second) while it hops between discharging/charging at 80%.
Even a “resume charging” setup where we can configure an allowable drop before charging again would go a long ways – I’ve had to hack my kernel to limit these notifications from getting tapped out to userland as frequently as they come in, because upowerd ends up pegging a core at the current rate and making the problem even worse.
That’s fine, and exactly how I’d like it, keeps the battery cycling low.
Agreed it helps to reduce battery cycling, but my understanding is that if it was to instead run off the mains rather than discharge the battery once the threshold is reached, the battery cycling would be lower. This seems desirable.
I also wonder if this would resolve the issue I get at the moment with firmware updates, in which I can’t update without discharging the battery a bit first (it will not update if the battery is being discharged, and at the moment it discharges at full charge).
I might not understand what you mean, but AFAIU the battery will also just discharge when plugged in at 100%. Even if it didn’t, it still seems beneficial from a battery longevity perspective if it just ran off the mains rather than discharging.
The problem being that the mains cannot, with the provided supply at least, be guaranteed to supply enough power under all circumstances, so the battery is used and secondly such a trickle charge does not create cycle life or heat the battery so any degradation ‘should’ be negligible. In fact normal time decay would be much greater.
If you would like an indepth analysis you may like to read
That’s interesting! I might misremember, but I think there’s a BIOS setting which allows the laptop to run off the mains directly. Does this mean that with this setting, it might use the battery still? Maybe it would be interesting to see whether the laptop will work in this mode without the battery.
Thanks for the link! I read through it, but it’s not really clear where this is suggested - were you thinking of a particular excerpt? I agree that small cycles are better than large ones, but I don’t believe that they avoid the problem entirely.
One conclusion I took from this was that one of the best things you can do to prolong the life of your battery (if you’re already doing a low depth of discharge, keeping within the appropriate temperature range etc.) is to increase the amount of time each cycle happens over - e.g. by only discharging it when necessary. I believe that running off the mains rather than discharging the battery would be one way of achieving this, and I seem to have gotten good results with other laptops that supported it in the past. Also I’d imagine that the manufacturers of these laptops (e.g. Dell) would probably have tested this charging mode and concluded that it was a worthwhile setting to include before shipping it with their laptop BIOSes, but I could be wrong.
If you are thinking of the battery disconnect feature this is so you don’t have to physically unplug the battery for maintenance, the system will power off after applying this setting, the battery will be “connected” again after connecting the power cable.
I have run my system (11th gen i5) with the internal battery disconnected with the Framework 60W power supply which works fine if you are not stressing the system but I believe this is why the recommendation is to use a 100W supply when running without battery. Under full CPU/GPU load my system drains ~3W from the battery but is this the problem you were describing?
I found this too: BIOS guide - #11 by Kieran_Levin which suggests similarly. Hopefully they will consider supporting this as more than a one-off use feature.
I don’t mind if the battery is drained under full load, my understanding though is that the battery is just getting discharged once it reaches its charge threshold (rather than the laptop just running off the mains). I could be wrong though! For example running acpi -i, I can see “charging at zero rate” when I’m at the battery threshold. Maybe this is the “hybrid mode”, where it will just use the battery when needed.
