How can I help with thinkpad x220 issues?

[Erich Dollansky]

Hi,

On Tuesday, July 03, 2012 08:51:33 PM Ian Smith wrote:
> On Tue, 3 Jul 2012 18:20:22 +0700, Erich Dollansky wrote:
>  > On Tuesday, July 03, 2012 03:34:36 PM Natacha Porté wrote:
> [..]
>  > > > How did you get this value? No battery inserted? The range should be
>  > > > from 0 to 100.
>  > > 
>  > > Indeed, I had no battery inserted. Years ago I took the habit of
>  > > removing the battery of laptops used for long amount of time on AC
>  > > (currently for my X220, that's 9h every workday), because keeping a
>  > > fully charged battery on AC used to kill it. However I admit I don't
>  > > know whether technology improved enough to mnake it a non-issue (or even
>  > > whether even by then it was actually an issue and not an urban legend).
>  >
>  > As an engineer I would say that Lenovo would have done a real bad job 
>  > if this makes still a significant difference.
> 
> Nonetheless, it's worth discharging and recharging both NiMH and Li-ion 
> batteries periodically.  If you do it once a week that's only 52 cycles 
> per year; a small fraction of its design cycles, and unless you store it 
> partially charged in a fridge, less usage than its 'shelf life' anyway.
> 
this is true.

> Apart from not harming the battery - which is designed for daily cycling 
> over 2 or 3 years - running it down past exhaustion now and again will 
> recalibrate the battery's onboard coulomb counter to reflect capacity.
> At least when they were called IBMs, that was IBM's advice for Li-ion.
> 
I did not know this but wondered how they do it.
> [..]
>  > > > > hw.acpi.battery.time: -1
>  > > > > hw.acpi.battery.state: 7
>  > > > 
>  > > > I have 0 here.
>  > > 
>  > > I have 2 when it's charging, and 1 when it's discharging.
>  > 
>  > Ok, now I have the same. 1 and 2. But never 7. It seems that there 
>  > are some secrets.
> 
> No secrets in FreeBSD, just stuff you have to hunt for and figure out :)
> 
> /usr/src/usr.sbin/acpi/acpiconf/acpiconf.c suggests hunting thus:
> 
> smithi on t23% find /sys/ -exec grep -H ACPI_BATT_ {} \;
> [.. see usage in acpi_battery.c, acpi_{cm,sm}bat.c and acpi_machdep.c ..]
> /sys/dev/acpica/acpiio.h:#define ACPI_BATT_STAT_DISCHARG                0x0001
> /sys/dev/acpica/acpiio.h:#define ACPI_BATT_STAT_CHARGING                0x0002
> /sys/dev/acpica/acpiio.h:#define ACPI_BATT_STAT_CRITICAL                0x0004
> /sys/dev/acpica/acpiio.h:#define ACPI_BATT_STAT_NOT_PRESENT     0x0007
> /sys/dev/acpica/acpiio.h:#define ACPI_BATT_STAT_MAX             0x0007
> /sys/dev/acpica/acpiio.h:#define ACPI_BATT_UNKNOWN      0xffffffff /* _BST or _BIF value unknown. */

Interesting. I did not get the idea. I must watch for the critical once the battery is low.

>  > > > > $ sysctl dev.acpi_ibm
>  > > > 
>  > > > I do not have anything with IBM or Lenovo.
> 
> Is this a Thinkpad model that shows no benefits from loading acpi_ibm?

I just loaded it to veryfy. There is no difference between loaded and not loaded. At least it does not do any harm. The sysctls also do not show up after being loaded.

Erich