Funny battery values (nx6325)

[Ian Smith]

On Mon, 15 Mar 2010, Joerg Wunsch wrote:
 > As Garrett Cooper wrote:
 > 
 > > > I think the 2 A are realistic. With a 5100 mAh capacity, it should
 > > > result in somewhat more than 2 hours of run time (maybe more if the
 > > > laptop eventually goes idle, and reduces CPU speed).
 > 
 > > If you were pulling 2 A, that's be the same approximate amount as
 > > many 2 x 1U servers.
 > 
 > Remember, this is measured at 10.8 V (or whatever the actual value is)
 > battery voltage...  The laptop is running on batteries here, how
 > should it be able to measure the mains voltage/current?  (Despite, the
 > mains part of the supply is completely decoupled by the PSU.)

That's right, 2A * 11V = ~22W which sounds perfectly reasonable.  My T23 
on battery draws about 24W @1133MHz idle, about 12W @733MHz, which I've
verified as my house supply is 12V, and on AC these draw metered 3.0 and 
1.5A respectively, ie 36W and 18W @12V, allowing inverter inefficiency.

 > This is what I'm getting on an old TP600E machine:
 > 
 > dhcp208# acpiconf -i 0
 > Design capacity:        34560 mWh
 > Last full capacity:     14080 mWh
 > Technology:             secondary (rechargeable)
 > Design voltage:         10800 mV
 > Capacity (warn):        1728 mWh
 > Capacity (low):         345 mWh

Which are 5% and 1% of nominal capacity, commonly used values.

 > Low/warn granularity:   1 mWh
 > Warn/full granularity:  1 mWh

Same on my T23.  cf below to your nx6325 values.

 > Model number:           ThinkPad Battery
 > Serial number:  
 > Type:                   LION
 > OEM info:               IBM Corporation 
 > State:                  discharging 
 > Remaining capacity:     99%
 > Remaining time:         1:18
 > Present rate:           10641 mW
 > Voltage:                11850 mV
 > 
 > 10.6 W / 11.85 V = 0.9 A  (the machine was idle here)

Looks about right to me.

 > No idea why the Thinkpad is returning Watts instead of Amperes, as
 > most other ACPI BIOSes do.

Same on the T23.  Likely depends on both the EC and the in-battery chip, 
but there must be an order of magnitude miscalculation 'somewhere'?

Requoting your original:

 > The battery is declared as 55 Wh, which would correspond to 5.1 Ah
 > (probably 3 x 2 x 18650 cells).

Sounds right.  My T23 battery is rated at 43200mWh design capacity.

 > When unplugging the AC supply, I get:
 >
 > remi# acpiconf -i 0
 > Design capacity:        279 mAh
 > Last full capacity:     279 mAh

These are obviously completely silly, and should be more like 5100mAh as 
you observed.  ie, they're something like 1/18 of their proper value, 
perhaps 1/20?

 > Technology:             secondary (rechargeable)
 > Design voltage:         10800 mV
 > Capacity (warn):        14 mAh
 > Capacity (low):         3 mAh

These are 5% and 1% of the silly value above.

 > Low/warn granularity:   100 mAh
 > Warn/full granularity:  100 mAh

About 9mWh, further showing how crook the capacity values are.
  
 > Model number:           Primary
 > Serial number:          00784 2006/10/04
 > Type:                   LIon
 > OEM info:               Hewlett-Packard
 > State:                  discharging
 > Remaining capacity:     98%
 > Remaining time:         0:08
 > Present rate:           2064 mA
 > Voltage:                11870 mV

This shows the remaining time calculation using these wrong capacity 
values.  0:08 times ~18 = 146m = 2:26 which sounds much more likely.

What happens if you let it discharge; how long do you really get?

cheers, Ian