I've got a new (third-party) battery for my trusty Thinkpad X201. I've used it for a few days and often the laptop just turns off hard (between 20...40% state of charge). At first I thought it to be a one-time defect and got a new one, but the problem stayed. Strangely, it never happened with the original Lenovo battery.
I've tried to calibrate the battery with TLP on GNU/Linux, but it always ends at 1% with
Currently discharging battery BAT0:
voltage = 9108 [mV]
remaining capacity = 960 [mWh]
remaining percent = 1 [%]
remaining time = 4 [min]
power = -11284 [mW]
state = discharging
force discharge = 1
Press Ctrl+C to cancel.
Error: battery BAT0 was not discharged completely. Check your hardware.
During my recalibration trials, I've noticed a pattern to the hard turn-off as TLP allows to discharge the battery whilst connected to AC, preventing the "belly landings".
If I load the battery to 100%, unplug and then use the computer in one long session, it properly detects low SOC and goes into standby. That's how it should be.
My normal pattern of use is different: I charge the computer to 100%, unplug, use it for some time and then enable standby mode. It's in this mode for a few hours to days and then I resume using it. When using it and a SOC of 40%...20% is reached the battery fails (i.e. laptop turns off when not connected to AC, or, when using TLP, the battery SOC jumps to about 1%).
By now, I can reproduce those turn-offs. The reason seems to be that the battery controller (?) doesn't track SOC drain whilst the laptop in standby mode. Steps:
- Charge to 100% in power-on state
- Enter standby mode
- Leave it for some days
- Turn on: The battery always shows 100%. Let it run, SOC decreasing in a normal fashion.
- ...until the laptop turns off hard at 40...20% (depending on how long the standby phase was).
I'm at a loss here, until now I thought a controller inside the battery tracked the SOC reliably. But now it seems that in standby mode this controller is also put to sleep, so that it isn't able to track the decreasing SOC until wakeup, and thinks we're still at 100%. Might this be a buggy controller inside the third-party battery? As stated, I haven't had this issue with the original part.
I'm open for any insights on how to calibrate the battery or any ways to have a reliable SOC display again.