Sorry for the delay, Russell.

On Mon, Feb 03, 2014 at 11:42:20PM +1300, RussellMc wrote:
> On 3 February 2014 21:00, James Cameron <quozl@laptop.org> wrote:
> >
> > Source code for the 8051 EC for the XO-4 can be found here:
> > http://dev.laptop.org/git/users/rsmith/ec-cl4-mmp3/
>=20
> Thanks.
> I had a wander through the source code files to see if I could find
> the end of charge parameters and "what happened next" for the
> LiFePo4 battery, but without success. It may be in there
> somewhere. I found code or comments for rate of charge, voltage for
> transition from CC to CV and voltage for low current trickling up of
> grossly undervoltage cells.

I spoke with our battery charging guru during a brief contact visit
today, and he pointed me at:

- LIFE_VOLT_CV of 7.3V; the threshold above which CV mode is to begin,

- LIFE_TAPER_CURRENT of 120mA; the threshold below which the battery
  is considered full, and;

- LIFE_SYSTEM_VOLTAGE of 7.4V; the voltage that is provided during CV
  mode.

After CV the system disconnects the charger from the battery, and does
not maintain voltage or current.

> What I was looking for was what decisions had been made on when to
> terminate CC phase and what voltage (if any) was then applied to the
> battery, and at what voltage charging would be reinitiated.

Check.  Check.  Ah, reinitiation.  We assume we are in full control of
the battery, and since it is removable we use the EEPROM on the gauge
to maintain a state of charge.  The gauge already has an accumulator.

Reinitiation of charge cycle occurs if either:

- the state of charge is not greater than 95%,

- the battery voltage is not greater than 6.150V.

The accumulator has insufficient bits, and can roll over if the
battery is discharged outside the laptop, so it is rare but possible
to see an inconsistency between state of charge and the accumulator.

> I have it on what should be good authority (and *no *direct personal
> experience) that the majority of the Chinese solar LiFePO4 charging
> designs have been deficient in this area and have been grossly
> damaging LiFePO4 batteries in consequence. As OLPC has had LiFePO4
> for some years and would presumably 'have noticed' if battery packs
> were grossly missing the 2000 cycle target I was interested to see
> what decisions had been made.
>=20
> It would be interesting to hear from the OLPC battery charging gurus
> re what decisions had been made for LiFePO4 en of charge termination
> and restart. It actually seems "simple enough" to get it right, but
> I do not (yet) have the in-field several year results that prove me
> correct.  Presumably OLPC do, and in large volume.

One of the other things we have is logging of the power system in the
operating system deployed.  CSV files showing the charge history and
laptop use have been instrumental in design of later models.

--=20
James Cameron
http://quozl.linux.org.au/
--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.