Re: [PATCH v3 01/18] dt-bindings: power: battery: add constant-charge-current property

From: Liam Breck
Date: Wed Feb 15 2017 - 15:22:00 EST

On Wed, Feb 15, 2017 at 12:53 AM, Quentin Schulz
<quentin.schulz@xxxxxxxxxxxxxxxxxx> wrote:
> Hi,
> On 15/02/2017 01:46, Liam Breck wrote:
>> On Tue, 14 Feb 2017 10:40:55 +0100 Quentin Schulz wrote:
>>> This adds the constant-charge-current property to the list of optional
>>> properties of the battery.
>>> The constant charge current is critical for batteries as they can't
>>> handle all charge currents.
>>> Signed-off-by: Quentin Schulz <quentin.schulz@xxxxxxxxxxxxxxxxxx>
>>> ---
>>> added in v3
>>> Documentation/devicetree/bindings/power/supply/battery.txt | 2 ++
>> Is constant-charge-current dependent on the battery (e.g. capacity, nominal voltage, etc) or the
>> system (charger chip, input current/voltage, etc)?
>> It belongs in Doc.../power/supply/battery.txt if it's a characteristic of the battery.
>> Note, this page asserts that constant-current charging applies to NiMH batteries:
>> Related properties to be added to battery.txt near-future in a patchset for the BQ24190
>> charger are as follows. These are not currently in enum power_supply_property, so the actual names
>> are still to be decided.
>> precharge-current-microamp:
>> maximum charge current during precharge phase (typically 20% of battery capacity)
>> termination-current-microamp (or endcharge-current):
>> a charge cycle terminates when the battery voltage is above recharge threshold,
>> and the current is below this setting (typically 10% of battery capacity)
> We have a client with a board whose battery accepts a maximum of 300mA
> for charging. So depending on the battery, we cannot have any charging
> current we want. The AXP PmMICs set constant charge current in a range
> of 300mA-1800mA, so it is enforced by the charger but needs to be
> adapted depending on the battery present in the system.
> The AXP PMICs charge battery with constant current (Ichrg) between the
> trickle voltage (Vtrkl which is ~3.0V) and the targeted voltage (Vtrgt;
> which seems to be the voltage telling the battery is fully charged).
> So if I understand correctly, "my" constant-charge-current would be
> located in the charging cycle between your precharge-current-microamp
> and the termination-current-microamp as it is the current for the
> charging process as a whole.
> See here[1] for the explanation in the datasheet (page 20).
> That would definitely match what is explained in your link for constant
> current.
> [1]
> Let me know if something seems odd,

Sounds OK to me. I'm happy to ack this, after my patchset goes in.
I'll be posting v7 this weekend, with a minor addition to battery.txt.

BTW there's rather a lot of ppl and lists CC'd on this, none of which
are listed in the patch comment...

> Thanks,
> Quentin
> --
> Quentin Schulz, Free Electrons
> Embedded Linux and Kernel engineering