Maker Pro
Maker Pro

Battery pack A31-x101 repair attempt of asus x101 eeepc


Feb 2, 2013
Feb 2, 2013
Hello together,
To introduce myself, first I´m an electronics engineer and hobbyist since >20 jrs now working in Hardware fail analitics for money. In long gone days cell change of Batt packs was easy. Nowadays some unknown obstacles - for safety I know - were built in the quite expensive Battpacks. With the Help of this Forum it might be possible to bring some light in the misterious dark of the

Battery of an asus x101 model e.g. a31-x101. that is a 3 cell one.

Mine had just 30 minutes power left after 9 months powerusing
The prices for this type of batt ranges from ~40 - 90 Eur.
This was enough for me to get interrested in repairing, better changing the cells.

My Debian based OS ( I am using Knoppix ) operated eeepc x101 terminal said when using
>cat /proc/acpi/battery/BAT0/info
just some 800 mA last full capacity.

So I decided to open the Batt-Pack and check the malicious cell.
Then I connected a junsi cell-logger with some soldering and found the middle cell more than 400mV difference to the right one. Battery detached
Battery connected + Adaper connected, the cell-logger told me the laptop-charging algorithm is not balancing the cells.
So i connected the removed batt pack to a separate lipo balancer charger.
Still all cells left soldered to the packs electronic in the case.
A retrial with
>cat /proc/acpi/battery/BAT0/info
after offline balancing told 1670mA last full capacity.
That was feeled too optimistic due to just some minutes more now without adapter.
So I made the descision fatal.
I disconnected the cells from the battpacks electronic.
And changed the cells with some i know are good and of same capacity.
And so bricked the batt.
I could have known better, but I did not

they describe a lot of things I did not know before desoldering.

In case someone has a working but still bad battery pack I would highly reccomend not to disconnect any of the wires from the cells to the controller. Change the cells as in an open Heart operation to keep the safety-stuff alive. see some more about this here

well now it is to late.
The pack lies around with fully charged cells and no voltage is comming from the + terminal
still some information can be retrieved from the pack. but it remains cut off.
The voltage reading is exactly the one the cell-logger reports.
But the serial number is missing now, and the cycles is set to 0

So far i have no idea which batt controller is really inside nor how to make the reset.
The Pack is opened in a way that I can not look at the electronics to identify something.

Meanwhile I got a new pack cause i want to be mobile again.
But I want to bring the battbrick back to life but how?

here what is left on the bricked Batt
>cat /proc/acpi/battery/BAT0/state ; cat /proc/acpi/battery/BAT0/info
present: yes
capacity state: ok
charging state: charging
present rate: 0 mA
remaining capacity: 0 mAh
present voltage: 12063 mV
present: yes
design capacity: 2600 mAh
last full capacity: 1670 mAh
battery technology: rechargeable
design voltage: 10800 mV
design capacity warning: 10 mAh
design capacity low: 5 mAh
cycle count: 0
capacity granularity 1: 26 mAh
capacity granularity 2: 26 mAh
model number: X101
serial number:

>acpi -b
Battery 0: Charging, 94%, charging at zero rate - will never fully charge.

and this is what the "new" battpack says
cat /proc/acpi/battery/BAT0/state ; cat /proc/acpi/battery/BAT0/info 
present:                 yes
capacity state:          ok
charging state:          discharging
present rate:            930 mA
remaining capacity:      1097 mAh
present voltage:         10772 mV
present:                 yes
design capacity:         2250 mAh
last full capacity:      2113 mAh
battery technology:      rechargeable
design voltage:          10800 mV
design capacity warning: 10 mAh
design capacity low:     5 mAh
cycle count:		  0
capacity granularity 1:  22 mAh
capacity granularity 2:  22 mAh
model number:            X101
serial number:            
battery type:            LION
OEM info:                ASUS

What was measured looking on the battpack terminals
Multimeter / Scope used (n1 vs x6 means GND of meter at n1 connectpin ,other at x6)

- n1 n2 x3 x4 x5 x6 x7 p8 p9 +

n1 vs n2 <1ohm (negative connector with <1ohm to the cells)
n1 vs x3 <1ohm
n1 vs x4 1Mohm
n1 vs x5 125Kohm
n1 vs x6 2,2V 50ms 0V 200ms repetitive, some glitches ( assume clock signal SCC)
n1 vs x7 0V (assume SerialCommunicationData for SMBus SCD)
n1 vs p8 0V
p8 vs p9 <1ohm

I tried to connect x4 to x7 over 100 ohm with gnd. no change.
I have an old parallelport equipped laptop, would it be worth to try the I2C stuff
exierience, discussion and help is welcome.