Something that has been going on with my L5 for months and months now, but which I have just lived with, because it’s not all that inconvenient…
1_ If I plug it in while it’s powered on, the charging LED does not come on, the battery indicator does not indicate charging, and the battery continues to drain.
(I haven’t tried waiting to see if it eventually activates, because I don’t want the battery to drain too far.)
2_ If I then power down the phone, charging and LED kick in after a minute or two. I can also power up again at this point without affecting the ongoing charging.
3_ If I plug it in while it’s already powered down, see item 1 above.
Neither I have some clue on why (quirky since recently) we see Librem 5 registers energy and energy-full values above 20 Wh? Anyway, @amarok, I do think that those upower -dmade-up-front values might cause within OP described charging behavior.
The energy-full value drifts all the time for me with my charging behavior. After a few days (I charge pretty frequently, also with the Lapdock 360) my energy-full is usually some 3-4Wh above my energy-full-design. Seems like shutting the device off, removing and reinserting the battery and restarting is a good weekly routine to reset the energy-full value. Annoying and quirky, but works for me.
No, that’s completely unrelated. The only thing it may influence is accuracy of reported percentage value. It’s based on gauge’s heuristics and if you keep plugging the phone for short charge cycles, it’s going to naturally drift away and there’s nothing wrong with that. Let it discharge until it shuts down and then charge it to 100% and it will likely get closer to the actual value.
Also, remember that the values you have shown are already transformed by upower. Use cat /sys/class/power_supply/max170xx_battery/uevent to see the unprocessed values reported by the gauge.
Yes, unrelated (although not easy for me to understand), thanks for helping me! Especially after I’ve successfully tested your very kind upgrade of Jumpdrive (archive.zip) everything looks fine now (and I’m content with your contribution again):
It’s odd, but I can live with it. As I said, maybe charging would eventually start in the 1st scenario situation, but I haven’t followed through to see if it would. (Yet.)
