Comment 1 Andre Klapper 2010-02-11 14:30:07 UTC

Thanks for reporting this.

Are you using the original Nokia charger or something else?
If you have another compatible charger, does the problem occur also with that? 
Can you please start X Terminal and provide the output of the commands:
* cat /sys/devices/platform/musb_hdrc/mode 
* cat /sys/devices/platform/musb_hdrc/charger
Can you please provide a syslog as described on
 http://wiki.maemo.org/Documentation/devtools/maemo5/syslog ?

REPRODUCIBILITY:
4/4, battery-eye running, shows a nice decreasing line from "Full" (95%) to
when I pick up my phone first time before I left for work.
http://dostortugas.org/download/screenshot01.png

Original Nokia AC-10E 1200mA 5V Charger (shipped with)

root@Nokia-N900-42-11:~# cat /sys/devices/platform/musb_hdrc/mode
b_idle
root@Nokia-N900-42-11:~# cat /sys/devices/platform/musb_hdrc/charger
1

I am charging now, klogd and sysklogd running, after that I will drain battery
and retry with charging over USB port (separated power input on usb-hub to get
it fast), I will provide you with syslog for both once I have done that. I am
9h travelling by train tomorrow, anything else to track?

I did not install battery widget or something else till I found out I was
starting of with less power than "full charged"

syslog
http://dostortugas.org/download/syslog_Feb_11_2010_2045

6/6
With both, usb and wall-charger hal doesnt do an update
I tried it now with a reboot while on usb and get
"battery.charge_level.percentage = 96  (0x60)  (int)" before
"battery.charge_level.percentage = 65  (0x41)  (int)" after

Having a look in the over time steps I get something like
96% to 78% with wall-charger plugged in within one hour and 96% to 65% with usb
in two hours.
These measurements were taken with charging from "OK" state waiting for green
LED "Full" and then start with power eating tasks to speed it up (transfer GBs
over wlan)

syslog
http://dostortugas.org/download/syslog_Feb_12_2010_0316
reboot after "Feb 12 03:01:42"

Comment 5 Andre Klapper 2010-02-15 19:49:16 UTC

Thanks!

Comment 6 Andre Klapper 2010-02-16 12:36:28 UTC

Hi Rüdiger, what is "reporting.design" and "reporting.current"?

> "battery.charge_level.percentage = 96  (0x60)  (int)" before
> "battery.charge_level.percentage = 65  (0x41)  (int)" after

How did you get these numbers? From lshal?

Comment 7 Andre Klapper 2010-02-16 15:15:30 UTC

Forwarding internal questions:



> EXPECTED OUTCOME:
> Green LED for telling "Fully Charged", capacity_state 'Full' and at least 100%
> Battery Charge Level or about the same current for reporting.design and
> reporting.current

What does "same current" here mean?

> ACTUAL OUTCOME:
> Green LED for telling "Fully Charged", capacity_state 'Ok', <93%(hi, 82%low)
> charge level, >90mAh(low, 180mAh hi) reporting.current & design difference (
> (charge level jumped from 94 to 83% within 10 minutes with nothin on but GSM,
> guess it was reading the first values from an old lshal not being triggered
> when I looked the first time, thats for when device was idle)

I can't understand this sentence (especially what is meant with "hi" and "low")
and I don't want to start guessing what this all means.

> alarm keeps the light on if started from shutdown device, dont know if that
> happens with idle device too.

I don't understand this sentence, please explain.

> With both, usb and wall-charger hal doesnt do an update

Doesn't do update on what situation?

> I tried it now with a reboot while on usb and get
> "battery.charge_level.percentage = 96  (0x60)  (int)" before
> "battery.charge_level.percentage = 65  (0x41)  (int)" after

Before what? After what?

(In reply to comment #7)
> > EXPECTED OUTCOME:
> > Green LED for telling "Fully Charged", capacity_state 'Full' and at least 100%
> > Battery Charge Level or about the same current for reporting.design and
> > reporting.current
> 
> What does "same current" here mean?

something like 1270 for design and 1258 for current

> > ACTUAL OUTCOME:
> > Green LED for telling "Fully Charged", capacity_state 'Ok', <93%(hi, 82%low)
> > charge level, >90mAh(low, 180mAh hi) reporting.current & design difference (
> > (charge level jumped from 94 to 83% within 10 minutes with nothin on but GSM,
> > guess it was reading the first values from an old lshal not being triggered
> > when I looked the first time, thats for when device was idle)
> 
> I can't understand this sentence (especially what is meant with "hi" and "low")
> and I don't want to start guessing what this all means.

[hi,low]; hi = highest value,low = lowest value
this all means that lshal was just useless to look for precise values, is it?
it also shows that I get a green light even with battery state on 'OK' level...

> > alarm keeps the light on if started from shutdown device, dont know if that
> > happens with idle device too.
> 
> I don't understand this sentence, please explain.

device was turned off and an alarm set, the device was with plugged wall
charger and the screen backlight didn't turn off automatically so it drained
battery to a serious state (if looking for a charged device for a working day),
means it was on 84% battery within first 10 minutes after boot with
wall-charger plugged in all night and while booting (set to stby, unplugged,
drove to work, checked battery {84%}).

> > With both, usb and wall-charger hal doesnt do an update
> 
> Doesn't do update on what situation?

didn't update any values within "lshal |grep batt"

> > I tried it now with a reboot while on usb and get
> > "battery.charge_level.percentage = 96  (0x60)  (int)" before
> > "battery.charge_level.percentage = 65  (0x41)  (int)" after
> 
> Before what? After what?

Reboot?! (are you guys serious? if I 'do' something and tell a 'before' and
'after' statement, it is probably 'before' and 'after' I 'did' something)

*** This bug has been confirmed by popular vote. ***

At the picture https://bugs.maemo.org/attachment.cgi?id=2849 I just attached to
another bug you see what I mean in the far right, green bottom boarder means
charger is connected.
Due to the bug this is attached to -  the drain is that high that you can
clearly see the issue. As this got an internal bug already I just update the
version to current.

Created an attachment (id=2987) [details]
Lithium ion discharge charts

I added li ion discharge chart as an attachment.

First of all depleting battery while charging only stops trickle charging. It
was good thing to do with NiMh but not with Li-ion (wears li ion cells quickly)

And secondly, those percentages are only sophisticated guesses as someone could
presume when looking at the attachment.