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

[mskala at ansuz.sooke.bc.ca]

On Fri, 11 Sep 2015, rsdio at audiobanshee.com wrote:
> Terminating a line involves adding a path to ground and/or a path to the
> power rail - i.e. a shunt to DC. You can't terminate a line with a
> series resistor. Actually, you can terminate a line with a series
> resistor if the final load is referenced to ground, but in the case of
> an opto-isolator there is no reference. Termination has to allow the
> current to be diverted so that it does not reflect back to the source.
> But termination is about voltage signals. With current loop signals, I
> don't think that termination even comes into play. If it did, then you'd
> need something other than series termination.

I don't understand this comment and I wonder how it can possibly be right.

A transmission line is two conductors.  The line has a characteristic
impedance, which will ordinarily be a pure resistance.  It is terminated
if, at the end of the line, there is a resistance (or something that looks
like a resistance) equal to the line's characteristic impedance.  If that
is the case, then at the start of the line, sources will see just the
line's characteristic impedance and not something more complicated.  With
a properly terminated line, signals will go from the start to the end and
then be absorbed nicely in the resistance.  In any other case, some part
of the signal will bounce off the end and propagate back to the start,
and the impedance seen by the source connected to the start of the line
may be something other than the line's characteristic impedance, and
other than a pure resistance.

Pretty often, especially if it's an unbalanced line, one of the conductors
will be connected to ground at some point.  But that has very little to do
with termination.  The termination is between the two conductors of the
line, nothing to do with DC ground.

At least, that's how termination works with the transmission lines used
in radio electronics.  Is this term used to mean something completely
different in the audio realm?  It wouldn't be the first time.

Are you just referring to a single conductor as a "line" and assuming
the other conductor is the unspecified path back through ground?  In that
case you'd certainly be right that such a path is necessary, but I
wouldn't call it a transmission line in the first place, and there'd be no
realistic hope of figuring out what the characteristic impedance actually
might be.
-- 
Matthew Skala
mskala at ansuz.sooke.bc.ca                 People before principles.
http://ansuz.sooke.bc.ca/