On May 9, 10:32 am, Laszlo Lebrun <
lazlo_leb...@laszlomail.com> wrote:
> On 09.05.2013 16:08, -hh wrote:
>
> > So you're claiming that you can insert*any* USB cord into your device
> > and have it be charged?
>
> No, the other other way around:
> The USB cable can be plugged into any USB *socket* available to charge.
> With non-Apple devices, even that cable is now standardized to micro-usb.
Ah, so you're focused in on the max power ratings of various USB
outputs.
> But the real point is that every USB socket can deliver enough power to
> charge reasonably any smartphone in 4 to 6 hours.
A Standard USB port is only required to deliver 1 unit (5V * 100mA) of
power.
With some devices having upwards of a 2000 mA-hr battery pack, it is
simplistically easy to say that this is 4 hours @ 5V & 100mW equals
2000mW and thus should be the charge time. However, this requires
assumption of zero overhead factors and a 100% end-to-end system
efficiency. The next layer of the onion would be to account for the
known major losses in the budget, such as the charger's electronic
efficiency (80-90%), the battery chemistry transformation efficiency
(~85%) and overhead of the charger's control system (50mW?), etc.
> Excepted Apple devices?
> Is that normal or has Apple included a circuit to *prevent* that?
Not necessarily. It has already been shown that Apples will actually
charge (albeit slowly) on a minimal connection. What's more likely is
that there's a much higher overhead expenditure which is preventing
the (low) available power from going into the battery - - some of this
could be the trade-off convenience factor of the device automatically
downloading iTunes/email/etc when it had adequate core power, or it
could be "more" overhead budgeting going into battery management...we
do not really know.
If you wanted to try to reduce the variables, you could take a 'mostly
discharged' iPhone which has already done all of the applicable "phone
home" bits, disable all wireless connections and then manually turn it
off (hopefully to not re-wake) and see how well it does in 4, 6, 8,
etc hours. This would best be done on a USB host source which is
known to not tolerate devices sucking more power than the 100mW that
they've been allocated.
-hh