How to short two pins when both are LOW

Hi all,

Im working on a somewhat universal powerbank for cellphones (lt1374 switching regulator provides 5v fed to female USB). It seems that to get a fast charge (1-2A) on an iPhone it needs 2.8v on D- and 2.0v on the D+ line of the USB, certain Samsung tablets need 1.25v on both D+ and D- and others (I believe newer spec) simply need the D+ and D- lines shorted.

You can get these voltages by using voltage dividers but in order to simplify I'm currently applying these voltages through PWM pins on an Arduino and an RC filter to convert the PWM to different voltages (set using a button and a switchcase). For the Samsung this works fine and I presume also for the iPhone (cellphones recognise high charge capability) but where I am stumped is how to create the third option: shorting the D+ and D- pins . How would you go about this? Transistors... mosfet?

A MOSFET has an internal diode that might cause problems. Maybe an analogue switch like the 4066.

EDIT: The SN74CBTLV1G125 has a lower on resistance.


Rob

I very much doubt your simple filtering in the PWM would be at all effective in providing the necessary sense voltages on the data lines for starters. If anything, you would be better off simply using a number of data pins to set up the appropriate resistive dividers.

A single MOSFET cannot guarantee to be switched on both for high and low voltages; a 74HC4066 (lower on resistance) would probably be OK and could be used to do some of your voltage switching as well.

If you find that the resistance of a CMOS switch is too high, consider a small DIP reed relay.

DIP-ReedRelay.jpg

HI all,

Many thanks for all your feedback! Excellent!

The SN74CBTLV1G125 looks interesting as does the reed switch will certainly give them a go.

@Paul__B, the simple filter does the trick when trying to convince a Samsung tablet, when PWM'ing 1.25V it starts drawing 1A at 5V, 1.5-2A at 5.4V instead of the usual 500mA. If the switches don't work out, yeah, ill probably try to use separate voltage dividers but am a bit low on pins (Project is somewhat more complex, GPS, SD card, PSP joystick, Parallel LCD, LEDs)...

Just don’t forget to set the PWM to give out 0V across both caps before shorting D+ and D-.
Those 1uf caps can store enough energy to take out the N74CBTLV1G125 or the delicate reed relay contacts.

In terms of on resistance the relay is a no-brainer, but there are better switches as well, eg the DG2012, Ron of 1R.


Rob

To make it more foolproof and also to simplify the design, I'd use separate plugs for each phone-type.

Don't most phones with the exception of the iPhone use mini or micro USB these days?

Having thought about it a little more, I will probably go the DvdDoug way, multiple USB connections, but will have to work on limiting current or OR'ing when two devices are connected, potentially drawing up to 4A...

@UnoDUeTre: Yes, many micro and mini usb devices exist, but they are generally compatible with standard USB, a new issue is wether to support USB 3, need to read up on that one... Thanks again, especcially for taking the time to draw a new schematic!

EDIT: Just found this: http://www.ti.com/lit/ds/slvsby8b/slvsby8b.pdf, USB charging port controller, looks interesting aswell.

EDIT 2: Yes, the TI ic is the one! Job done.

fredkadet: Just found this: http://www.ti.com/lit/ds/slvsby8b/slvsby8b.pdf, USB charging port controller, looks interesting aswell.

That TI TPS2513 chip looks great. I'm planning on building a phone charging station, I think that chip would make life a lot easier. Do you know when to use the TPS2513 verses the TPS2514? Also in the example circuit they have a TI TPS2561A current limiting power switch. Would you really need this? If your power supply had sufficient power, then just let the phone draw as much current as it wants, right?

ScottG: That TI TPS2513 chip looks great. I'm planning on building a phone charging station, I think that chip would make life a lot easier. Do you know when to use the TPS2513 verses the TPS2514? Also in the example circuit they have a TI TPS2561A current limiting power switch. Would you really need this? If your power supply had sufficient power, then just let the phone draw as much current as it wants, right?

My thoughts exactly, great little IC! No more multiple USBs, voltage dividers etc... From what I've read the 2513 is just the dual usb port version of the single port 2514. A current limiting IC is probably a good idea when using the dual USB controller charging two connected devices, but i'll leave that aswell, my lt1374 can deliver over 3A and will only supply a single USB port.

I've ordered a couple and expect to be up and running in a few days. Will report back how they perform.