[sdiy] Possible MIDI interface powered by MIDI driving device...

[rsdio at audiobanshee.com]

On Mar 24, 2016, at 5:09 PM, Tom Wiltshire <tom at electricdruid.net> wrote:
> On 24 Mar 2016, at 20:24, Olivier Gillet <ol.gillet at gmail.com> wrote:
>>> Steve T's comment about 5V not being a reliable supply suggests to me that for things like this (let's call it "creative rule-bending"?!) we should use uPs that can operate from 5V down to 3V or so. There are such things, so this shouldn't be too difficult to organise.
>> 
>> The value of the resistors on the OUT side depend on the supply
>> voltage. If the MCU is powered by 3.3V, they should be much lower than
>> 220 ohm (about 50 ohm) - otherwise you won't be sending enough current
>> in the loop!
> 
> Ah, yes, of course. Good point, Olivier. I was only thinking of the IN side.
> 
> I suppose you could use the uP to detect its own power supply voltage and adjust the outgoing current accordingly, but at that point, it's all starting to get a bit more complicated than it's worth for a thing that removes 0xFE from a MIDI stream.

It is not typical for a uP to have adjustable control over the outgoing current from a GPIO port pin. It is fairly common for a uP to have some sort of A/D pin that could be used to measure the supply voltage, but turning that into a variable output current would require additional resources. In other words, I agree that this is getting a bit more complicated than it's worth for something that is desired to be "simple."

Brian