How long laptop battery should live ?

Ian Smith smithi at nimnet.asn.au
Wed Nov 24 04:55:23 UTC 2010


In freebsd-questions Digest, Vol 338, Issue 3, Message: 12
On Tue, 23 Nov 2010 22:34:16 +0100 David DEMELIER <demelier.david at gmail.com> wrote:

 > I just realized that my HP laptop lost the half battery capacity. Take a look :
 > 
 > markand at Melon ~ $ acpiconf -i 0
 > Design capacity:	4400 mAh
 > Last full capacity:	2132 mAh
 > Technology:		secondary (rechargeable)
 > Design voltage:		14400 mV
 > Capacity (warn):	200 mAh
 > Capacity (low):		100 mAh
 > Low/warn granularity:	100 mAh
 > Warn/full granularity:	100 mAh
 > Model number:		Primary
 > Serial number:		02109 2009/08/11
 > Type:			LIon
 > OEM info:		Hewlett-Packard
 > State:			charging
 > Remaining capacity:	0%
 > Remaining time:		unknown
 > Present rate:		3061 mA
 > Voltage:		15751 mV

'Remaining capacity: 0%' should indicate that you ran acpiconf -i0 when 
the battery was fully discharged, and you'd just started charging it?

Was that the case?  Seems a bit strange that the battery would already 
be at 15.75V at start of charge, though the 3A charge rate sounds fair.
Does it come up to showing 100% charged later?

 > Design capacity:	4400 mAh
 > Last full capacity:	2132 mAh
 > 
 > I'm guessing if this should be, the laptop has only about 8 months and
 > I don't use it much on battery, only around one day per week and not
 > until the battery is empty. If I remember correctly my old laptop
 > didn't lost so much of capacity in a few months like that.

Yes 8 (or 15) months is way too soon to lose half its capacity, even if 
you'd been running it on battery for a lot of the time.  This old Compaq 
laptop's battery (Li-ion, 14.4V) is at least 7 years old, and still runs 
for about 1.5 hours, ie 40-50% capacity, and in a subtropical climate.

 > My question is : how long a battery should live ? is anything could
 > bring down the battery sooner ?

It could just be a faulty one, but you'd be very lucky to get a battery 
replaced under warranty.  It says it's an HP battery, manufactured only 
15 months ago, so it's unlikely to be a 'shelf life' problem.

 > I hope the next battery wont be too much expensive and will live more..

It may be that the little chip inside the battery that counts charge in 
and out has lost track of itself.  This tends to happen more commonly on 
Li-ion batteries that are rarely or never fully discharged, and at least 
some manufacturers including IBM recommend 'conditioning' batteries from 
time to time to correct this.

For example, this year I bought two second-hand IBM Thinkpad T23s whose 
batteries both responded very well to about three full discharge / full 
charge cycles; one now shows about 2/3 of original capacity, up from 
about 1/3, and both batteries were likely already several years old.

To get to full discharge you need to run it right down, beyond where the 
OS would normally shutdown or suspend at somewhere between 5% and 2% of 
capacity.  What I do from that point is start it up in the BIOS Setup, 
turn off anything that would have it shutdown or suspend on low battery, 
and wait till it turns off, completely exhausted.  From the BIOS Setup 
screen there's no danger that this could mess up a filesystem.

Then plug it in and let it fully charge, preferably without running it. 
Once fully charged, turn it on and check its 'Last full capacity' again.  
It may take three or more such full cycles until you see no improvement.

While you're at it, you could actually time how long it runs till fully 
discharged; this also may improve somewhat, but what we're correcting 
here it the battery's own BELIEF in its capacity and state of charge.

HTH, Ian  (please cc me on any reply, I take this list as a digest)


More information about the freebsd-questions mailing list