[ltp] X240: Sudden battery death

Henrique de Moraes Holschuh linux-thinkpad@linux-thinkpad.org
Mon, 28 Dec 2015 10:41:53 -0200

On Tue, 08 Dec 2015, Jochen Spieker wrote:
> Jeffrey L. Taylor:
> > Quoting Jochen Spieker <ml@well-adjusted.de>:
> >> 
> >> I usually use the laptop on AC, but every or every other week I need to
> >> unplug it anyway and then try to discharge at least the external battery
> >> completely. Unfortunately, I cannot tell how many charging cycles it had
> >> because cycle_count returns 0 for some reason.
> >> 
> > 
> > Why are you discharging the external battery completely?  If it is NiCad
> > (Nickel Cadmium) this is resonable behavior (the memory effect).  If it's a
> > lithium battery, this will kill it quickly.
> "Completely" as in "down to 5%, then the internal battery takes over". I
> don't do a deep-discharge, of course.

5% is cutting it too close to deep discharge for Li-ION, especially because
you cannot really trust it to be 5% (and not, say, 3% or less) as the "fuel
gauge" can accumulate a lot of error very quickly.

Keep it above 20% if you can.

> As I said, I am aware that this is not the best way to treat a battery.
> But I am still surprised by it becoming almost useless in less than two
> weeks.

Well, yes, that steep device breakdown curve is quite surprising.  If it is
a 3-cell or 6-cell battery, it looks like each cell-group suffered a
breakdown, close together in time...

But if cycle_count returns zero, it might well be battery pack
microcontroller failure (there's one inside the battery pack that is *NEVER*
powered down -- so, it is the one thing that must not have software bugs

  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh