On 4/6/2013 4:35 PM, Mark Zenier wrote:
In article , Francis C. wrote:
William Sommerwerck wrote:

I believe the resistor is needed to "clue" the power supply that
something that "wants" to be charged is connected.

I'm confused because I just tested a "normal" cable and this was
my observation:

Motorola RAZR V3 connected to wall-wart USB charger -- fail
Motorola RAZR V3 connected to automotive USB charger -- fail
Motorola RAZR V3 connected directly to Linux PC -- charged!

The weird thing is that I used the same cable for all 3 tests!

USB has power control from the host. The Linux system is going to
interrogate the phone and set up the internal hub/controller to whatever
power level the phone wants.

On a Windows box, the Motorola phone driver isn't built in. It looks
like Linux includes it so you don't need to install Motorola Phone Tools
in order to charge.

But I think you've got it backwards. The USB power controller on the
computer does not allow setting different current levels as requested by
the phone.

What really is happening is that the USB driver is telling the phone
that it's plugged into a USB port and that the phone can charge at
500mA. If you trick the phone into charging at 1.25 amp by using a
resistor in the cable then the over-current protection on the USB port
will probably trip (it doesn't trip at 501mA, more likely to trip at
800-1000mA, but 1250mA is probably going to trip it). If you trick the
phone into charging at 500mA with a resistor then it would work fine.

On my Asus/Google Nexus tablet, if you short the USB data pins then it
thinks it's plugged into a 2.1A charger no matter what it's plugged
into. You need to do this to use a dumb 2.1A to 12V to USB adapter, and
it's part of the USB spec. But it will charge from a Motorola Micro USB
car charger too, just slower.

USB ports were not designed to be used as chargers, but the USB 3.0 spec
has addressed the charging issue pretty well.