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

[Richie Burnett]

> Interestingly, the standard shows a single 220R resistor in the input
> circuit, although I have seen several manufacturers (e.g., Roland)
> split that into a 100R and a 120R resistor, one each side of the
> opto-isolator.

Very strange, there must be some reason for that.  There's no way they'd 
spend an extra half cent on every unit manufactured without some tangible 
benefit.  (Maybe the product already used loads of 100R and 120R resistors, 
but didn't use a 220R anywhere else!?)

> For EMC/RFI/noise reasons I would recommend have equal source
> impedances on the two lines to maintain a semblence of balanced
> impedances.

Good point.  I hadn't thought about impedance matching / termination issues. 
With today's micros being much faster than when the MIDI standard was 
conceived it's interesting to note that no slew-rate limiting seems to be 
needed on the UART output going to the MIDI port in order to ensure 
compliance with radiated emissions standards.  I'm guessing that the 220R 
resistor does a good job of slugging the rise and fall times of the voltage 
on the MIDI cable itself.

-Richie,