Powering USB device from TI 5V?

[+jedimatt42]

I'm a programmer, my electronic skills are mostly paint by number, but I've managed to get a Teensy 3.1, USB Host shield (mini variety) and some taps into the TI itself to translate a USB keyboard press into the letter A.

 

Some pics

 

 

https://goo.gl/photos/xhLKCTwCPuAeUbP86

 

My trouble is with power... I've tapped the internal power supply +5V, and ground, and it will power the whole mess including the USB keyboard, however, it won't boot the USB keyboard. So in the picture you see more juice coming in from a PC into the teensy's programming port.

 

Does anyone have advice about how to get the power needed out of the internal power supply? I don't see anything around that states availability of current.

 

Is there a wise way to tap the power supply? In particular, when I need more power at start up?

 

-M@

 

 

 

[Tursi]

hmm.. have you tried measuring the current draw? I never measured mine, but I had no trouble powering an AVR and PS/2 keyboard off the internal power supply? I tested a lot of keyboards, including a couple of old AT ones, and they were all fine. I'd expect a USB keyboard to draw comparable current to PS/2.

[+jedimatt42]

The keyboards vary, many have hubs built in... the one I want to use is rated at 100ma, the one I've been testing with is rated at 40ma... But your advice to measure is completely sound. And, it'll give me a solid reason to get a multi-meter that does that.

[Tursi]

Yeah... I'm just wondering if something with the wiring is drawing more than expected... that Tiny should be drawing less than my old ATMega too, but maybe they've tied some of the power lines in a funny way. (Alternately, how are you powering the the Tiny? Is it a Trinket that I see there? I did a project recently with one myself, and I ended up feeding the power in a little oddly to bypass some of the regulation circuity on the board.)

 

Either way, quickly eliminated as a suspect if the current draw is as expected.

[+jedimatt42]

Ok, good news is that this wasn't a power problem... It turned out to be a software problem. I started with an example 'sketch' for the USB Host library, that had a wait forever loop in it for the Serial logging connection to be established before it would initialize the USB system.

 

Just a tad embarrassing.

 

I measured the current, starting at the source, which was only about 45mA and didn't spike at the beginning, as I had assumed.

Then I measured the current just at the USB keyboard, and it was very low 6mA. (cause it wasn't even being given the go ahead to boot)

 

Anyway, without the blocker in the code, the whole thing powers up nicely.

 

I did review my wiring and found a ground connection they recommend that I left out. So, win, win, win...

 

Thanks.