[maemo-users] N900: Force charging on USB?

From: Nils Faerber nils.faerber at kernelconcepts.de
Date: Wed Apr 14 13:58:27 EEST 2010
Joerg Reisenweber schrieb:
> [Nils Faerber Di  13. April 2010]:
>> I am preparing for a vacation and want to avoid carrying too many
>> accessories.
>> So I took my universal USB charger (which is just a USB plug carrying
>> the +5V) plugged in the Nokia micro-USB cable and connected to the N900.
>> Well, it does not charge - in contrast to almost any other device.
> Probably it does, just it charges at 100mA only, and it doesn't signal it's 
> charging.

Well, pitily it does not - I just checked with a "USB charger" which has
a dual color LED turning red on current flow. When I connect the N900
via the normal USB cable there it does not change in any way. Just
connecting the charging adapter change the LED from green to green plus
a little red which mean current flow (just from the regulator).

So I think we can safely assume that a blank USB cable will not in any
way charge, even not at 100mA.

>> This is I think also perfectly valid since there is no host at the other
>> end to negotiate the charging current.
> No, according to USB specs any source of VBUS 5Volt USB has to deliver 100mA. 
> And the charger chip enables charging @ 100mA when detecting external 5V on 
> USB - completely compliant with these specs.

Yes, sure. What I meant is, if the charger is designed in a way to only
properly work when sucking more than 100mA it would be valid to refuse
charging. If it could handle a 100mA slow charge but it apparently does
not do that I would see this as a bug.

>> So assuming I know what I am doing is there a way to enforce the N900 to
>> start charging even if there is no host connected? This must work
>> somehow since the power adapter presumably does not contain a USB host ;)
> Exactly. The charger shorts D+ and D- data lines of USB port.
> This should enable fast charging

I will try that - press thumbs that it works ;)

>> Some fiddling with /sys/... or DBus would be fine with me...
> Alas we got no sysfs nodes (yet ;-D ) to control the BQ24150 USB charger chip 
> directly. BME does all the 'magic' for now, and afaik there's no API to tell 
> BME to switch to a different charging mode.

Umpf... too bad.

The D+/D- trick sounds promising, many thanks!

> cheers

kernel concepts GbR        Tel: +49-271-771091-12
Sieghuetter Hauptweg 48    Fax: +49-271-771091-19
D-57072 Siegen             Mob: +49-176-21024535
More information about the maemo-users mailing list