@Kieran_Levin
You mentioned that the current sense resistor might not be accurate, thus the 85% limit being applied by the EC.
Would you be able to tell us the component part number for the current sense resistors used in the FW13 and FW16, so that we can improve our measurements.
For more than 10A current, using resistor is not a good way because of the waste power and the added voltage drop (the latter makes the battery to drain even when the load is not maxed), the use of hall effect sensors is highly recommended.
using a value lower than 100% assumes that the tolerance is downward, i.e. actual resistance is lower, this leads to higher actual current at a given sensed current.
However, real world testing concludes that this is not the case.
I think Iām in the same boat as many others here and have a problem related to battery power management. Yesterday, I played Shadow of the Tomb Raider for a while, and after 25-30 minutes, the laptop started throttling extremely (sound crackling, FPS dropping below 10 frames per second, the whole system becoming extremely slow). After I managed to shut down the laptop and restart it a few minutes later, the problem persisted. Even testing the RAM takes up to a minute instead of a few seconds.
The problem first occurred when I set up the device with Windows to check the hardware for errors. Under FurMark, the entire system slowed down, and at first, I thought it was overheating. However, the temperatures only reached a maximum of 85Ā°C on the CPU and GPU/VRAM.
ā¦ Something else Iāve noticed with Manjaro, which Iām now using instead of Windows, is that when I plug in the power adapter (240W USB-C charger), the laptop switches between charging and AC power every second. This is indicated by the LED flashing on and off. If I unplug the adapter and wait a moment, everything is usually fine.
What do I need to do to change this behavior? I have the Framework 16 with RX7700S GPU module.
FYI upower regressed again in 2026, the battery symbol alternates between battery and battery with AC plug symbol when charged to the set limit
Thanks! Not happening on my Arch system but I havenāt updated everything for a while. Iāll hold off on update upower then.
Let me add what Iāve experienced for completenessā sake.
My config:
- OS:
Gentoo Linux - Machine:
Framework 13with7840U CPU - UEFI: Latest firmware as provided to me through the stable channel of LVFS.
- Bought in Januari 2024. Have been running for 2 years without any issue.
Started encountering this for the first time today, after I had my latop unplugged for a little under 2 hours, wherein my battery charge dropped to 77%.
Then after it charged back up, the cycling started with the battery at 100%. It also doesnāt matter if the system is under load, even with all cores at 100%, it still cycles.
- Power supply always present according to
/sys/class/power_supply/ACAD/uevent - LED near charge port never off, have always used the Framework 60W PSU that came with it.
- Battery state toggles between: āDischargingā and āNot chargingā as seen in
/sys/class/power_supply/BAT1/uevent.
Donāt know if I have provided any new information with this (havenāt read the entire thread), but hopefully my ānever having this problem for 2 years and it then suddenly appearingā is also useful info.
I would also like to add I did not do any updates in the past few weeks or so, so nothing shouldāve changed programmatically about the system, which would lead me to believe the issue might at the very least be partially hardware related, or perhaps be influenced by battery age/condition.
Finally my machine is usually on/not-suspended and powered through USB-C. I have a tendency to use it more like a desktop than a laptop, except for it also being a desktop you can carry. Current uptime 15 days.
In case itās of any interest to people here, I had ChatGPT make a small script that automatically detects the 180W / 240W chargers and then apply 100% rating input charge limit override. Iām a software engineer, and both tested the code and reviewed the code, in case anyoneās sensitive about not trusting the AI.
1 Like
Just for the record and @Matt_Hartley - opening a support ticket doesnāt do anything. Iāve opened a ticket almost 2 months ago, and there has been 0 feedback from support
I didnāt have any fantasy that it was going to change anything. Coming to 5 years. A multi-generation issueā¦ The talk about support ticket is just, IMO, BS at this point.
This needs action, result. ā¦and thereās neither.
Itās almost like Burton with their Step On:
Leaving their community disappointed.
It would be interesting to hear if anyone else with the battery flip problem and an Intel 13gen updated to the 3.09 firmware and has similar issues as I have:
Ever since I updated, the battery status became almost useless. Reporting absolutely insane numbers in short bursts