I got a Framework 13 two years ago (Apr 2024), and the battery just died. Like, dead dead. Spicy pillow dead. I’ve had tons of laptops over the years, but I’ve never had one where the battery died that quickly.
I didn’t even notice it until the fan stopped working today, and I took it off the dock to check it out, and noticed the input panel was bulging up from the chassis. I think that bulge may have pinched the fan, which could explain why it wasn’t spinning, but the sensor reported it was spinning at several thousand RPM.
I opened up the chassis and every cell of the battery is bulging. Kinda scary, considering it’s been sitting on my desk at 100% charge for a few days now.
It is generally best to keep the battery at around 70% if you mainly use it while plugged into power.
Only charge it to 100% when you know you will be using it and discharging it soon.
There are settings in the bios to do this charge limiting.
Lithium ion battery life is shortened if the battery is left 100% and if left 0%.
Contact FW support.
They might replace it for you as they had a batch of bad ones.
Sensor issue, the fan control uses closed-loop feedback. The temperature determines the RPM and the duty cycle is controlled to reach that. You may be able to start the fan by commanding a duty cycle directly using fw-ectool’s $ sudo ectool fanduty 100
So, the battery protection only kicks in if you’ve had it plugged in for more than a few days, and I just plugged it in like two days before, so it was still in the 100% charge window.
This issue is common. Mine swole and the mousepad needed way more force to click. But it still worked and charged fully. Support sent me a free replacement battery.
I contacted support, and they said that since it’s out of warranty, I have to buy a new battery. So I guess the takeaway here is that if you buy a Framework, you might have to buy a new battery for it every two years.
Considering their products are already on the higher end in terms of price, this really leaves a very sour taste in my mouth. I would expect the product they send me to be of high quality, and if not, I would expect them to at least try to make it right.
Framework Support followed up clarifying that they extended the warranty on the bad batteries by one year, and mine is one month out of that extended warranty, so f*ck me I guess.
Not that I was planning to buy any additional Framework products after it came out that they financially support DHH, a white supremacist, but I feel like now I’m going to be an advocate against Framework.
I think we are getting closer to the cause of the battery issue. In addition to voltage that probably makes the 55Wh more durable than 61Wh. The main problem is very likely to be BMS calibration or programming issue.
Firstly, the charging temperature is too high, probably due to Celcius to Kelvin conversion error. The battery only stops charging at 55C but slows down to a crawl as early as 20C
It would be better for the battery health if the temperature window is moved from 20 - 55C to 10 - 45C.
Secondly, the battery charging termination current is way too low. Near the end of charging, the voltage is held constant while the current decreases. When the current is lower than a threshold, the current turns off and charging is finished.
Most laptop I have used charges from 99% to 100% in 5-10 minutes. Even for laptops that I haven’t used for months (higher BMS error that needs a full charge to calibrate), 30 minutes is the most from 99% to 100%.
In the Framework Laptop, charging from 99% to 100% may take 30 minutes to 2 hours even if the battery has been calibrated by charging to 100% yesterday, showing that the charge termination current is way too low. This is also a contributing factor to battery aging.
Now we find the culprit, the next step is for Framework stuff to relay the message to the battery manufacturer to adjust the BMS accordingly.
I had some more messages with the support team after installing a new battery I had to buy:
I’ve now received the new battery and installed it, and the laptop is struggling to charge. I’ve tried four different adapters, and four different USB-C cables, and the laptop keeps disconnecting and reconnecting the charge. Like, it makes the noise like you just plugged in the charger, and the charge icon goes from discharging to charging and back over and over. Is that normal with a new battery?
So, there were more issues that just the battery. None of the USB ports that had non-USB C expansion cards (a USB A and an HDMI) would work, and with the USB C ones there was that constant disconnect/reconnect when charging. I booted up into a Kubuntu live USB, and they all worked again. So I copied my entire disk to a USB drive and tried reinstalling Fedora. Now everything works again and the charging issue is gone. I’m not sure what happened, but maybe the system constantly having issues when it got too hot (it would remount the disk as read only when the NVMe got too hot) broke something in the system. Anyway, it’s all working correctly now after the reinstall, so I think we’re good.
I’ve now put a hard limit on the battery charge in the BIOS to 85%. It sucks that I don’t get that extra 15% of battery, considering the battery life is already pretty terrible, but I’d rather have to find an outlet more often than have to buy a new battery every two years.
One really good thing that I think has come out of Framework being around is that the Apple MacBook Neo seems to prioritize repairability. That’s probably a direct response to Framework. Hopefully that’s a signal that the industry may be shifting.
I can’t edit the old post so I guess I’ll reply here. @James3 Sorry I made the wrong conclusion from the EC program you found.
Charging at high temperature is not likely to be the reason of excessive wear because the battery stops charging at 45C, not 55C, according to my latest test. However, the long 99%-100% (low charge termination current) is still a problem.
I played a video game on battery so high discharge rate heated it up. Then I plugged in an 100W charger and stopped the game so I expected most power went to the battery but I got this.
$ sudo framework_tool --smartbattery
Enter unseal key in hex (e.g. 04143672), or press enter to skip:
Device Name: FRANGWAT01
Manuf Name: NVT
Serial Num: 0035
Manuf Date: 2023-08-21
Chemistry: LION
FW Version: Device=0x0045 FW=09.03 Build=0x4900
Mode: 0x6001
Temperature: 47.8C
Voltage: 14.838V
Cell 1: 3.710V
Cell 2: 3.709V
Cell 3: 3.712V
Cell 4: 3.708V
Current: 0.000A (avg 0.000A)
Charge: 21% (689 / 3337 mAh)
Design: 3915mAh @ 15.480V
Bat Status: 0x00C0 [DSG (Discharging), INIT (Initialization)]
Cycle Count: 346
=== Battery Health Analysis ===
Status: HEALTHY
No issues detected. Battery is operating normally.
Summary:
Cycle count: 346
Capacity: 3337 / 3915 mAh (85%)
Current cell balance: 4mV spread
$ sudo ectool chargecontrol
Charge mode = NORMAL (0)
Battery sustainer = on (75% ~ 80%)
$ sudo ectool chargestate show
ac = 1
chg_voltage = 15480mV
chg_current = 0mA
chg_input_current = 4500mA
batt_state_of_charge = 21%
When the battery was cooled down to below 45C
$ sudo framework_tool --smartbattery
Enter unseal key in hex (e.g. 04143672), or press enter to skip:
Device Name: FRANGWAT01
Manuf Name: NVT
Serial Num: 0035
Manuf Date: 2023-08-21
Chemistry: LION
FW Version: Device=0x0045 FW=09.03 Build=0x4900
Mode: 0x6001
Temperature: 44.0C
Voltage: 16.169V
Cell 1: 4.051V
Cell 2: 4.039V
Cell 3: 4.043V
Cell 4: 4.036V
Current: 3.894A (avg 1.769A)
Charge: 21% (693 / 3337 mAh)
Design: 3915mAh @ 15.480V
Charging: 3.915A @ 17.600V
Bat Status: 0x0080 \[INIT (Initialization)\]
Cycle Count: 346
=== Battery Health Analysis ===
Status: HEALTHY
No issues detected. Battery is operating normally.
Summary:
Cycle count: 346
Capacity: 3337 / 3915 mAh (85%)
Current cell balance: 15mV spread
$ sudo ectool chargecontrol
Charge mode = NORMAL (0)
Battery sustainer = on (75% ~ 80%)
$ sudo ectool chargestate show
ac = 1
chg_voltage = 17600mV
chg_current = 3912mA
chg_input_current = 4500mA
batt_state_of_charge = 22%
