0

Correction Notice

I'm sorry to bring this issue up based on a misremembering the cell count in series. There were not 4 cells in series, but 3 cells. Silly me. For 3 cells in series, the maximum battery voltage should be 4.2V * 3 = 12.6V, which seems quite consistent here, so "voltage calculation part" of the question is inaccurate.

Original Question

I'm using LXDE and battery applet shows 100% as battery state. Other commands supports that percentage:

ceremcem@cca-erik:~$ acpi
Battery 0: Full, 100%
ceremcem@cca-erik:~$ upower -i `upower -e | grep bat`
  native-path:          BAT1
  vendor:               LGC
  model:                DELL 49VTP27J
  serial:               7849
  power supply:         yes
  updated:              Fri 17 Jan 2020 11:46:50 AM +03 (105 seconds ago)
  has history:          yes
  has statistics:       yes
  battery
    present:             yes
    rechargeable:        yes
    state:               fully-charged
    warning-level:       none
    energy:              73.26 Wh
    energy-empty:        0 Wh
    energy-full:         73.26 Wh
    energy-full-design:  48.84 Wh
    energy-rate:         0.0111 W
    voltage:             12.014 V
    percentage:          100%
    capacity:            100%
    technology:          lithium-ion
    icon-name:          'battery-full-charged-symbolic'

Actual problem is that the nominal battery voltage is 14.4V (3.6V/cell, 4 series). That means at 100% SOC (state of charge), the battery should be at 4.2 * 4 = 16.8V. If you see the battery -> voltage section, it is currently at 12.0V, which means 3.0V per cell, which means it's at 0% SOC (fully empty) according to batteryuniversity.com:

enter image description here

How does those commands decide the SOC? Are they directly asking to the battery? How can we re-calibrate this mismatch?

ceremcem
  • 2,351

1 Answers1

0

This is typically done through SMBus, refer to http://smartbattery.org/specs/sbdat110.pdf for more information. So yes, this gets asked directly to the battery pack.

  • Is voltage also being asked over SMBus, or is it directly measured by the computer hardware (as it is quite possible)? – ceremcem Jan 17 '20 at 10:05
  • @ceremcerem a quick glance at the SMBus spec makes me think it's asked over SMBus, too. (e.g. page 24 of the spec) – Renan Jan 17 '20 at 10:55
  • 1
    @Renan Yes, it seemed like so. However, as the voltage measurement is a relatively easy task, I thought that a computer might handle this with its own instrumentation layer for redundancy purposes. Although it's nearly impossible to decide the SOC by looking at the battery voltage for Lithium based batteries, this might be a redundancy for cut off point decision, for both low and high limits, as a safety measure. So, it seemed like a worthwile question to ask. – ceremcem Jan 17 '20 at 11:40