Talk:XO LiFePO4 Recovery Procedure

From OLPC

Jump to: navigation, search

Contents

A dangerous experiment

dont know where i get ot here but ok ill try:

well folks, just after a glance at the vast detail of the recovery procedure guys i suppose i am right to post right HERE:

what i did:

i opened the lid of the batterie pack and after a while i did it:

i changed the two lifepo4 for 4 standard liion 18650

and ok, i runs...

cant well tell -normally the capacity should be DOUBLE!!

can someone of the hard/software guys enlighten us what about "smart batterie" status of the xo1 <-> batteries

is there ANY??

i prefered to SWITCH SUCH COMPLETELY OFF -if POSSIBLE

(and yes i know i then have to watch ot about the voltage etc.. not to talk about danger explosion, fires ... blah blah...)

but the advantage is:

more power, use of ANY used batteries from scrap, as there are a lot out there worldwide...!!!

if you want i can make pics of my mod... allows single batterie change in a second, lid closes well, these two mechanisms are well suited for!!!

and: yes sirs, did same at my eeepc, too: works, too!! , thomas@biopilze.de

VERY bad idea!

The XO charging system is not designed to charge the Li Ion batteries found in most laptops. Not only will it not effectively charge Li cobalt batteries its extremely dangerous. You have created an extreme safety hazard for yourself. Mis-charging lithum colbalt batteries will result in explosion.


(reasked by thomas@biopilze.de: please explain: what is the problem? out of my experience with older 18650s LiIon-batteries seem to withstand even higher voltages (happened about 4.4v) without problems (yet)... and: i had only put these older 18650-s loosely into the batterie-slot in the xo, this results in bad contact -so NORMALLY the cells just WONT CHARGE completely... -SO WHERE IS THE D A N G E R man??? -CANT SEE it YET!!!, thomas)

It's not the voltage. The charging circuitry will over-charge a lithium-ion battery. Imagine trying to charge a NiMH with your EEE pc: they're different chemistries

i've an other problem

my battery did not charge full, so when i try to run bat-recover to fix that, i get the response: no response from battery see-bstate responde: ff and xo hangs up watch-battery gives "normal" information...

any ideas?

Bizarre Results

I'm having some issues with my battery, and I'm not clear on whether the recommendations on this page apply to me or not. My xo shuts down when the battery claims to be at about 40%. I hoped that resetting it with batman's bat-set-low would help, but that appears to have made no difference.

I was hoping to shine some light on my problem, but olpc-pwr-log is giving me weird readings for net ACR. I was under the impression from this page that net ACR shouldn't be in numbers over 3,000, but it changes by thousands per reading! I'm figuring that, for some reason, I'm getting net ACR in units other than those described in this page, but I don't know what units. Can anyone shine some light on this?

A LiFePO4 battery mod that makes sense

This page attempts to describe a way to rebalance a battery pack that has become unbalanced. When discharging the cells (running on battery) the XO is sensitive to the output voltage of the battery pack. When charging cells, the charger is sensitive to the resistance/current of the battery pack. If one of the cells is at 80% capacity and the others are nearly dead, the output voltage will drop and the XO will shut down. If you plug it in, the charge will flow to the batteries and the cell that started at 80% will be full when the dead batteries have only charged to 20%.

Keeping the XO battery load in a closed circuit when it is running off external power would deep-discharge the battery, allowing the 80% cell to discharge through the other cells that are mostly discharged. This causes the low cells to discharge to lower voltage. If any of the low cells bottom out, they will not accept a charge and the whole pack will be dead. The solution is to "bottom balance" the cells when the pack is low but not dead: connect all of the cells in parallel[1]. Connected in parallel, the highest cells will discharge and the lowest cells will charge until they all have equal voltage. The cells can all be reconfigured in series and charged normally, and they will accept a charge up to the capacity of the lowest capacity cell. When they discharge, their voltages will all approach the minimum voltage at almost the same time, and will be mostly balanced. The capacity of the pack will approach the sum of the capacity of the cells as the cells are closer to balanced.

If we could design a circuit that reconfigures the battery pack in parallel when there is no voltage or just above the rated cell voltage, maybe that could be added to the battery pack?

Personal tools
  • Log in
  • Login with OpenID
About OLPC
About the laptop
About the tablet
Projects
OLPC wiki
Toolbox