View Single Post
      11-12-2012, 03:06 AM   #12
Lieutenant Colonel

Drives: 2009 Crimson 328i
Join Date: Oct 2012
Location: Los Angeles

iTrader: (0)

So I did a lot of testing on this today. What I found was interesting. First, I'm quite sure that the 'accessory doesn't support charging' problem is caused by an under-voltage to the phone. The source of the under-voltage can be a bunch of different things. First, BMW put the USB interface box in the trunk, meaning there's a few feet minimum, maybe 10 or so, of USB cable between the MULF and the port in your armrest. That length means lots of resistance on the power pins, which means lots of voltage drop when the device plugged in is drawing lots of power. A thumb drive might take only 50 or 100mA tops, while your charging phone can draw more than an amp if it's allowed.

So why did I investigate and what did I do? I noticed that my phone would charge fine without complaint from the Y cable through my lightning adapter, or from the normal white USB cable plugged in directly. BUT when I plugged either in through a 6' USB extension I got (to extend the Y cable out to a distance useable with the phone mounted on the dash), either cable would cause the phone to complain.

I plugged in my USB analysis cable (I made this, it just breaks out all 4 USB pins in the middle, you can turn on/off whether each is connected, and you can hook meters to them to measure voltage and current. Basically, I found:

With the extension cable, voltage before the lightning cable dropped to 4.5 or so under load. THIS caused the phone to complain. WITHOUT the extension cable, it was a bit higher, at 4.6 or 4.7V depending on current draw (the phone was at a relatively high charge at this point). What's interesting is that in all of these cases, the phone drew 200mA or so.

The most interesting thing is, with the phone deeply discharged (30% or so), WITH the 6' extension and with the y+adapter cables, the phone DIDN'T complain about charging, and drew 280mA. Voltage was decreased, as you'd expect (higher drop across the few meters of cable between the MULF and the phone).

I think the bottom line comes down to the same that is the case charging an iPad from a normal USB port: The iDevice will sometimes complain about charging not supported based on what REAL power it's drawing out of the cable. BUT, in ANY case, it's always drawing power if it needs it, and you can probably just ignore what it says. That said, if it complains that charging isn't supported, it probably means it's getting little enough power that it'll only charge with the screen off/not using lots of LTE/GPS/etc.

One final note: how much power the phone CAN draw is dependent on the current it IS drawing and the voltage it is GIVEN (i.e. it WILL draw 100mA, but the voltage drop on the cable means the phone sees something DEFINITELY LESS than 5V). How much power the phone TRIES to draw is dependent on what the host controller TELLS IT IT CAN. It seems that the MULF never awards the iDevice even 500mA, so unfortunately, no matter what USB power injector you connect, you WON'T EVER get the kind of high-speed charge you'd expect being plugged into a 1A or 2A wall or car charger.

The problem THERE is that the iPhone only knows it's allowed to draw >100mA if there's some kind of pre-arranged voltage being held on the data lines to tell it it's plugged into a dumb charger. You can't have those voltages and the active data communication with the MULF at the same time.

THAT SAID: the CableJive solution should be able to work: the USB lines on the 30-pin connector are never used (the Y cable actually has a USB->RS232 converter in it which talks to the serial pins on the 30 pin connector) so in theory, you should be able to present the dumb-charger voltages on those pins while passing the serial pins through from the Y cable to the iPhone (provided, obviously, you source power that will be sufficient to satisfy the demands you've just told the iPhone it's allowed).