[sdiy] MIDI phantom power...over 5 pin MIDI connector ?

[rsdio at audiobanshee.com]

On Sep 10, 2015, at 3:04 PM, "Richie Burnett" <rburnett at richieburnett.co.uk> wrote:
> Thanks for clarifying that you can just change the resistor values to make the current right to make MIDI work with 3.3V logic.  I did wonder whether you still needed a +5V line just to tie up pin 4 of the MIDI output jack.
Just remember that the lower limit is 2.8 V, which would require 0 ohm resistors. Below 2.8 V, you wouldn't be able to get 5 mA to flow through a standard MIDI input.

> Since MIDI is a current loop standard, I should in theory be able to implement a MIDI output port by grounding pin 5 of the DIN socket, and switching a current-source in the high-side connected to pin 4 of the DIN socket then?  That would still meet the standard in terms of an on/off modulated 5mA current loop and drive the floating opto-coupler at a MIDI input port just fine.  (I wonder if anyone does this?)
I had the same idea for my first commercial MIDI product back in 2008, but I found that the 5 V charge pump was cheaper. There are current source chips, and discrete designs that could get the job done, but I never found a compelling design.

Keep in mind that a current-source still needs to provide enough voltage to induce a current flow. I have a hunch that this would be more complex than simply generating 5 V and using the simpler circuit.


> I'd imagine that it would mess up any product that anticipated the "Electrical Specification Diagram" example shown in the MIDI spec and planned to draw power from pin 4 though?
I personally think that the MMA should call out those MIDI phantom power devices as non-compliant and be done with them. I think that's basically what they've done, although at the same time they also explain how to do it anyway.

Brian