Electronics question about a PDA cradle.

[xmastree]

One for the electronics experts here...

I recently bought a charge/sync cradle for my PDA. It was a cheap one from Hong Kong, bought on eBay.

http://cgi.ebay.co.uk/ws/eBayISAPI.d...m=120322181676

At first, I thought it was fine. Connects ok, and charges. But then I noticed that it only charges when the computer is on. So I investigated further.

With just the charger plugged in, it charges. With just the PC on and the USB plugged in, it charges. Both plugged in, PC off it doesn't, and the LED is dim.

It appears that the 5V from the charger is connected directly to the USB plug, and thus feeds back into the computer when it's off. I noticed this last night when I turned it off, a couple of the computer's LED's were dimly lit until I disconnected the cradle.

So, I'm thinking of a solution, and I have three options so far.

#1. Insert a Schottky diode in the positive line from the USB port, to block the reverse current but still allow charging in the absence of the 5V adaptor.

#2. Disconnect the 5V from the USB completely, relying solely on the charger for charging.

#3. Perform #2 with a relay, so that it will still work just with the USB connected, but plugging in the charger will switch over 5V from the USB to the charger. There is plenty of room inside, and maybe even another LED to indicate which source it's using.

My main question is, will the data connection be affected if the 5V isn't connected to it?
The 0V will always be there.

[Mach]

Not an expert (as you know by now ) but would like to help. To your question, I think so based on discussions over at

http://www.usbmadesimple.co.uk/ums_2.htm

https://www.usb.org/phpbb/viewtopic....ighlight=power
https://www.usb.org/phpbb/viewtopic....ighlight=power
http://www.design-reuse.com/articles...designers.html

Seems that Vbus (+5V) is used as a reference for D+ and D-. In addition, it appears that the USB host (PC) side interrogates Vbus draw to determine USB device status. At least that's my 15 minute interpretation.

Maybe ask at usb.org?

[Luke122]

Throw a switch on the 5v line.. easiest way to test.

They are all good ideas though, why not try them all?

[xmastree]

Seems that Vbus (+5V) is used as a reference for D+ and D-. In addition, it appears that the USB host (PC) side interrogates Vbus draw to determine USB device status. At least that's my 15 minute interpretation.

Thanks for the links, I'll investigate those. I do know that it's not quite as simple as one might think. I know that a usb device will tell the port how much current it requires, and the port will then limit itself accordingly. Which means that some non PC devices (like my older pda) wouldn't work if connected to a generic usb charger.

Seems rather strange to me, why not just give it 5V and let the device draw what it needs up to 500mA?

So, I think I'll experiment, and if it is true that Vbus is the reference, well I can always put the switch in the negative instead.

[ContraptionMaker]

My main question is, will the data connection be affected if the 5V isn't connected to it?

I might be a little late but do you have a USB extension cable? If so an easy to test is cut the 5v wires on your extension cable and plug the charger USB cable into it, then plug it into the computer. If you can still transfer data then you can disconnect it in the charger base. If not all you have ruined is a USB cable. If the USB cable can be unplugged from the charger then just try it with a regular USB cable. Again, can't hurt anything but the cable.

[XcOM]

i would recommend this, modify the cradle to have a switch, then use said switch the switch between the usb and the charger, then when you turn the pc off just switch it over to the charger, and as far as im aware the 5v+ isn't used for data, 0 and G are used for the data, 5v+ and 5v- are used to power devices.

But don't hold me to it, i haven't got my multimeter to hand