Battery recovery issues

Emiliano Pastorino epastorino at
Mon May 4 11:13:12 EDT 2009


I've attached batdbug.log and seebstate.log.
watch-battery says "No battery".
Battery led is always off.
Sn is: 00602080608110000873
bat-charge reports this:
320.83 mAh (7d53) 1428.12 mA (2ddc) 6.492 V (195c) Chg: 0.41mAh (  29)
then every column raises line to line (I copied that by hand because
bat-charge-log always says "Can't open file", even when usb stick is
plugged in)

If the first column is battery's charge, then it's almost dry.
Should I try bat-recover or charge it the usual way?

After all these tests, I'll have to write a step-by-step guide for the
people at our technical center so they can tell when a battery can
be recovered and when not, so I must try to cover as many situations
as possible.

I'll try bat-recover with a bunch of batteries today, so maybe tomorrow
or on Wednesday I'll be sending you some logs, if it is ok to you.

On Fri, May 1, 2009 at 3:29 AM, Richard A. Smith <richard at> wrote:

> Emiliano Pastorino wrote:
>  Richard,
>> I've just received a box with 60 faulty batteries inside, so I'll be
>> playing with them
>> for the next few years...
> :)
>  I did a "bat-recover" on one of them for about 18 hours and I noticed
>> this:
>> When I run "watch-battery", it still says "No battery". I did a full-reset
>> of the XO
>> but nothing happened, it still says "No battery".
>> Then I loaded batman.fth and ran bat-charge and I got a nice output. All
>> the values seemed to be OK when charging or discharging the battery.
>> I tried "batman-start; 6a bat-set-status; batman-stop" and I could see tha
>> 6a in
>> the first block, but "watch-battery" still says "No battery".
> Hmm.. and see-bstate shows 0 1 2 over and over?
> There is one more battery debugging tool available.  Its called bat-debug
> and bat-debug-log.with the power for the cpu and for the
> The both read the same thing but bat-debug-log will write the contents to
> 'disk:\batdbug.log'.  'disk' is USB or SD depending on what you have
> inserted.  bat-debug just does the screen and serial port.
> If the 1-wire state machine is just looping over and over bat-debug won't
> provide much more info.  It might however point out what part of the state
> machine is failing.  That part of the code has a pretty large number of if()
> clauses all lumped into the same state.
> Procedure:
> 1) Remove the problem battery.
> 2) Boot machine and stop boot at OFW prompt.
> 3) run bat-debug (or bat-debug-log)
> 4) insert the battery
> 5) let it run for one or 2 screenfuls of info
> 6) hit a key to stop bat-debug
> send me the info.  Note, that you don't run batman-start before you run
> bat-debug since you want the EC state machine to run and you don't need to
> 'fload batman.fth'.  bat-debug should be in your firmware already.
>  What is the difference between batman's "bat-charge" and "watch-battery"?
>> Tell me if I'm wrong, but I think that "bat-charge" reads battery info
>> directly from it,
>> and "watch-battery" takes that info from the EC. So, the problem could be
>> that
>> the EC isn't synced with the battery. Am I right?
> Correct.  Batman code takes over the 1-wire communication bus from the EC
> and talks directly to the battery.  'watch-battery' uses EC commands to read
> what the EC thinks.  So if the EC state machine is bailing out for some
> reason then you will get odd things from watch-battery where batman only
> needs the 1-wire to work.
>  So long I could recover 2 batteries out of 4. I'll try more batteries, the
>> batteries
>> that seem to be ok now are the same model (GP NTA2490), and the other two
>> (the ones I couldn't recover even with bat-recover) are
>  BYD LP183662AR-2S.
> I have no idea what this number is.  The serial number I would need is the
> long string of digits under the barcode in the center of the battery.
>  If you want me to do a particular test with any of these batteries, just
>> ask and I'll
>> share my results with you.
> Well.  I'd like to make sure the firmware has the diagnostics that will
> allow you to figure out whats up with the battery.  So depending on what
> see-bstate and bat-debug info is I'll perhaps need to make new firmware or
> new diags in batman.fth to try and figure out whats up.
> --
> Richard Smith  <richard at>
> One Laptop Per Child

Ing. Emiliano Pastorino
LATU - Plan Ceibal
Av. Italia 6201 CP: 11500, Montevideo, Uruguay
Tel: (598 2) 601 5773 int.: 210
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: batdbug.log
Type: text/x-log
Size: 8294 bytes
Desc: not available
URL: <>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: seebstate.log
Type: text/x-log
Size: 575 bytes
Desc: not available
URL: <>

More information about the Devel mailing list