[Thinkpad T410] Unable to discharge/recalibrate the battery

[cmonty14]

[x] I've read and accepted the Bug Reporting Howto

Describe the bug

Running discharge or recalibrate shows this error: Error: target level (0%) is too high compared to the actual charge level (98%) of battery BAT0.

Expected behavior

Discharging sets laptop to state "discharging", recalibrate starts procedure to calibrate battery.

To Reproduce

Steps to reproduce the unexpected behavior:

1. Put laptop in AC.
2. Run command sudo tlp discharge BAT0 or sudo tlp recalibrate BAT0

Additional context

❯ sudo tlp-stat -b -s
--- TLP 1.9.0 --------------------------------------------

+++ System Info
System         = LENOVO ThinkPad T410 2537EY8
BIOS           = 6IET85WW (1.45 ) LENOVO
EC Firmware    = 1.24
OS Release     = CachyOS
Kernel         = 6.18.2-1-cachyos #1 SMP PREEMPT_DYNAMIC Fri, 19 Dec 2025 14:28:57 +0000 x86_64
/proc/cmdline  = BOOT_IMAGE=/vmlinuz-linux-cachyos root=UUID=382d47e1-343d-48ab-99eb-331352a7bafd rw rootflags=subvol=@ loglevel=4 rcutree.enable_rcu_lazy=1 rd.luks.name=cb2b6cbf-799e-4f55-99f0-339db7660cc5=cryptroot root=/dev/mapper/cryptroot rootflags=subvol=@
Init system    = systemd 259
Boot mode      = BIOS (CSM, Legacy)
Suspend mode   = s2idle [deep]

+++ TLP Status
tlp            = enabled, last run: 10:08:15, 689 sec(s) ago
tlp-rdw        = enabled
tlp-pd         = not installed
Power profile  = performance/AC
Power source   = AC

+++ Battery Care
Plugin: thinkpad-legacy
Supported features: charge thresholds, chargeonce, discharge, recalibrate
Driver usage:
* tp-smapi (tp_smapi) = active (status, charge thresholds, force-discharge)
Parameter value ranges:
* START_CHARGE_THRESH_BAT0/1:  2..96(default)
* STOP_CHARGE_THRESH_BAT0/1:   6..100(default)

+++ ThinkPad Battery Status: BAT0 (Main / Internal)
/sys/devices/platform/smapi/BAT0/manufacturer               = SANYO
/sys/devices/platform/smapi/BAT0/model                      = 42T4763
/sys/devices/platform/smapi/BAT0/manufacture_date           = 2024-01-06
/sys/devices/platform/smapi/BAT0/first_use_date             = 2024-01-06
/sys/devices/platform/smapi/BAT0/cycle_count                =     87
/sys/devices/platform/smapi/BAT0/design_capacity            =  48840 [mWh]
/sys/devices/platform/smapi/BAT0/last_full_capacity         =  38700 [mWh]
/sys/devices/platform/smapi/BAT0/remaining_capacity         =  38310 [mWh]
/sys/devices/platform/smapi/BAT0/remaining_running_time_now = not_discharging [min]
/sys/devices/platform/smapi/BAT0/remaining_charging_time    =      3 [min]
/sys/devices/platform/smapi/BAT0/power_now                  =   7838 [mW]
/sys/devices/platform/smapi/BAT0/power_avg                  =   7888 [mW]
/sys/devices/platform/smapi/BAT0/state                      = charging

/sys/devices/platform/smapi/BAT0/start_charge_thresh        =     96 [%]
/sys/devices/platform/smapi/BAT0/stop_charge_thresh         =    100 [%]
/sys/devices/platform/smapi/BAT0/force_discharge            =      0

Charge                                                      =   99.0 [%]
Capacity                                                    =   79.2 [%]

I can run this to forcebly discharge battery:
echo 1 | sudo tee /sys/devices/platform/smapi/BAT0/force_discharge

[linrunner]

Hi. Please show the output of

sudo tlp-stat --psup -b 

[linrunner]

And right after that, please:

sudo sh -x /usr/sbin/tlp discharge

[cmonty14]

❯ sudo tlp-stat --psup -b
--- TLP 1.9.0 --------------------------------------------

+++ Battery Care
Plugin: thinkpad-legacy
Supported features: charge thresholds, chargeonce, discharge, recalibrate
Driver usage:
* tp-smapi (tp_smapi) = active (status, charge thresholds, force-discharge)
Parameter value ranges:
* START_CHARGE_THRESH_BAT0/1:  2..96(default)
* STOP_CHARGE_THRESH_BAT0/1:   6..100(default)

+++ ThinkPad Battery Status: BAT0 (Main / Internal)
/sys/devices/platform/smapi/BAT0/manufacturer               = SANYO
/sys/devices/platform/smapi/BAT0/model                      = 42T4763
/sys/devices/platform/smapi/BAT0/manufacture_date           = 2024-01-06
/sys/devices/platform/smapi/BAT0/first_use_date             = 2024-01-06
/sys/devices/platform/smapi/BAT0/cycle_count                =     88
/sys/devices/platform/smapi/BAT0/design_capacity            =  48840 [mWh]
/sys/devices/platform/smapi/BAT0/last_full_capacity         =  38700 [mWh]
/sys/devices/platform/smapi/BAT0/remaining_capacity         =  30780 [mWh]
/sys/devices/platform/smapi/BAT0/remaining_running_time_now = not_discharging [min]
/sys/devices/platform/smapi/BAT0/remaining_charging_time    = not_charging [min]
/sys/devices/platform/smapi/BAT0/power_now                  =      0 [mW]
/sys/devices/platform/smapi/BAT0/power_avg                  =      0 [mW]
/sys/devices/platform/smapi/BAT0/state                      = idle

/sys/devices/platform/smapi/BAT0/start_charge_thresh        =     20 [%]
/sys/devices/platform/smapi/BAT0/stop_charge_thresh         =     80 [%]
/sys/devices/platform/smapi/BAT0/force_discharge            =      0

Charge                                                      =   79.5 [%]
Capacity                                                    =   79.2 [%]

+++ Power supply diagnostic
/sys/class/power_supply/AC/type: Mains
/sys/class/power_supply/AC/online: 1
/sys/class/power_supply/BAT0/type: Battery
/sys/class/power_supply/BAT0/present: 1
/sys/class/power_supply/BAT0/capacity: 80
/sys/class/power_supply/BAT0/cycle_count: 0
/sys/class/power_supply/BAT0/energy_full_design: 48840000
/sys/class/power_supply/BAT0/energy_full: 38700000
/sys/class/power_supply/BAT0/energy_now: 30780000
/sys/class/power_supply/BAT0/power_now: 0
/sys/class/power_supply/BAT0/voltage_min_design: 11100000
/sys/class/power_supply/BAT0/voltage_now: 12263000
/sys/class/power_supply/BAT0/charge_behaviour: [auto]
/sys/class/power_supply/BAT0/status: Not charging
/sys/class/power_supply/BAT0/device/path: \_SB_.PCI0.LPC_.EC__.BAT0

+++ udev diagnostic
Reading rules file: /usr/lib/udev/rules.d/85-tlp.rules
  RUN{program} : /usr/bin/tlp auto
OK.

Command sudo tlp discharge is working as expected:

Currently discharging battery BAT0 to 0% (0 mWh):
voltage            =  12138 [mV]
remaining capacity =  30260 [mWh]
remaining percent  =   78.2 [%]
remaining time     =    105 [min]
power              =  17211 [mW]
state              = discharging
force discharge    = 1

And state switches correctly to /sys/devices/platform/smapi/BAT0/state = discharging.

So, this is pointing to a syntax error of my previously used commands sudo tlp discharge BAT0 or sudo tlp recalibrate BAT0. And I can confirm that sudo tlp recalibrate is working, too.

And it looks like using BAT0 fails because my device has only 1 battery.

[linrunner]

So, this is pointing to a syntax error of my previously used commands sudo tlp discharge BAT0 or sudo tlp recalibrate BAT0

That doesn't seem entirely plausible to me. sudo tlp discharge BAT0 is valid syntax and equivalent to sudo tlp discharge in your case (your battery is named BAT0).

If you can reproduce the behavior, I would be very interested in the output of:

 sudo sh -x /usr/sbin/tlp discharge BAT0

Via https://gist.github.com/ please.

[cmonty14]

I see...
Here's the relevant output of sudo sh -x /usr/sbin/tlp discharge BAT0.

Update:
After successful discharge my laptop rebooted. Since then, sudo sh -x /usr/sbin/tlp discharge fails with error. Here's the relevant output.

And just for completeness the content of some dirs in /sys/class/... and /sys/devices/...
Comparing this with previous run of sudo tlp-stat --psup -b the following lines are missing now. In my understanding this is resulting in error running sudo sh -x /usr/sbin/tlp discharge. Assuming this is correct, the question is: why are these files missing now?

/sys/class/power_supply/BAT0/energy_full_design: 48840000
/sys/class/power_supply/BAT0/energy_full: 38700000
/sys/class/power_supply/BAT0/energy_now: 30780000

❯ ll /sys/class/power_supply/BAT0/
-rw-r--r-- 1 root root 4,0K  5. Jan 15:28 alarm
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 capacity
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 capacity_level
-rw-r--r-- 1 root root 4,0K  5. Jan 15:27 charge_behaviour
-rw-r--r-- 1 root root 4,0K  5. Jan 14:49 charge_control_end_threshold
-rw-r--r-- 1 root root 4,0K  5. Jan 13:54 charge_control_start_threshold
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 charge_full
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 charge_full_design
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 charge_now
-rw-r--r-- 1 root root 4,0K  5. Jan 15:28 charge_start_threshold
-rw-r--r-- 1 root root 4,0K  5. Jan 15:28 charge_stop_threshold
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 current_now
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 cycle_count
lrwxrwxrwx 1 root root    0  5. Jan 15:27 device -> ../../../PNP0C0A:00
drwxr-xr-x 2 root root    0  5. Jan 15:28 extensions
drwxr-xr-x 3 root root    0  5. Jan 13:54 hwmon2
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 manufacturer
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 model_name
drwxr-xr-x 2 root root    0  5. Jan 15:28 power
-r--r--r-- 1 root root 4,0K  5. Jan 13:54 present
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 serial_number
-r--r--r-- 1 root root 4,0K  5. Jan 13:54 status
lrwxrwxrwx 1 root root    0  5. Jan 13:54 subsystem -> ../../../../../../../../../class/power_supply
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 technology
-r--r--r-- 1 root root 4,0K  5. Jan 13:54 type
-rw-r--r-- 1 root root 4,0K  5. Jan 13:54 uevent
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 voltage_min_design
-r--r--r-- 1 root root 4,0K  5. Jan 13:55 voltage_now

❯ ll /sys/devices/platform/smapi/BAT0/
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 barcoding
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 charging_max_current
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 charging_max_voltage
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 chemistry
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 current_avg
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 current_now
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 cycle_count
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 design_capacity
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 design_voltage
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 dump
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 first_use_date
-rw-r--r-- 1 root root 4,0K  5. Jan 14:49 force_discharge
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 group0_voltage
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 group1_voltage
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 group2_voltage
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 group3_voltage
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 inhibit_charge_minutes
-rw-r--r-- 1 root root 4,0K  5. Jan 14:49 installed
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 last_full_capacity
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 manufacture_date
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 manufacturer
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 model
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 power_avg
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 power_now
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 remaining_capacity
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 remaining_charging_time
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 remaining_percent
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 remaining_percent_error
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 remaining_running_time
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 remaining_running_time_now
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 serial
-rw-r--r-- 1 root root 4,0K  5. Jan 15:17 start_charge_thresh
-rw-r--r-- 1 root root 4,0K  5. Jan 13:55 state
-rw-r--r-- 1 root root 4,0K  5. Jan 15:17 stop_charge_thresh
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 temperature
-rw-r--r-- 1 root root 4,0K  5. Jan 15:29 voltage

[linrunner]

Your machine shows glitches when reading from /sys/class/power_supply/BAT0/energy_full.

When checking the first time, the file is not there L835. Then again in L871.

Interestingly, /sys/devices/platform/smapi/BAT0/last_full_capacity delivering the same value can be read. But I don't want to use /sys/devices/platform/smapi/BAT0/*, as there are reading glitches that I've already had to deal with (#793).

I am reluctant to include even more workarounds for this old hardware.

[cmonty14]

Gemini gives this explanation:
Even though TLP successfully reads your battery via tp_smapi (the "ThinkPad" way), the script has a secondary check that looks at the standard /sys/class/power_supply/ path. On your CachyOS/Kernel 6.18 setup, tp_smapi has successfully hidden those files to prevent conflicts. TLP's batdrv_discharge function sees the "missing" files, sets the capacity to 0, and then throws the error: "0% is too high compared to 80%." (It's mathematically nonsense, but it's how the script is written).

Sounds reasonable, but I don't know if it's true.

[linrunner]

Of course, I could make the error message more plausible.

Unfortunately, that doesn't help if energy_full cannot be read. A little later, energy_now cannot be read either (L881), and without it, the discharge process cannot be controlled.

[linrunner]

Well. I just noticed your edit above that /sys/class/power_supply/BAT0/energy_* are now completely absent. Instead there are charge_full, charge_full_design and charge_now which hints at an Embedded Controller (EC) issue.

Try to reset the EC as follows:

• Shut down
• Remove charger
• Remove battery
• Hold the power button at least 30 seconds

[cmonty14]

However, I cannot reset EC everytime a metric is missing.

The question is how often your EC has this problem. But anyway: in 2026, I won't be programming against the whims of hardware from 2010/2011.

In my understanding the battery is charging automatically when it is empty. And the laptop must not turn off.
But what if the laptop turns off and I need to boot manually? This would stop tlp recalibrate.

tlp recalibrate terminates as soon as the battery is empty as no program control is required for charging. So charging back to 100% won't be affected. tlp recalibrate has already set the thresholds to 96/100 in preparation.

Apropos "turn off". As a user of an Arch-based distro with the most recent software, you will encounter another problem: at 2% charge upower will shutdown the laptop – despite being connected to AC: https://gitlab.freedesktop.org/upower/upower/-/issues/312

It took quite some time to resolve this in upower, but there is no upower release for it yet.

Workaround in the current upower version is to either stop upower.service during recalibration or configuring CriticalPowerAction=Ignore in /etc/UPower/UPower.conf.

[cmonty14]

Well. I just noticed your edit above that /sys/class/power_supply/BAT0/energy_* are now completely absent. Instead there are charge_full, charge_full_design and charge_now which hints at an Embedded Controller (EC) issue.

Try to reset the EC as follows:

• Shut down
• Remove charger
• Remove battery
• Hold the power button at least 30 seconds

Is the absense of /sys/class/power_supply/BAT0/energy_* files related to an EC issue only?
I don't understand the requisites for these files to be created, either when system boots or after startup.

I tried 3x in the meantime to reset EC, but I don't get the files back.