My guess is some of you know the trick of using a "fake-balanced" cable
to connect an unbalanced output to a balanced input.

I mean using a say 50 ohm series resistor to pin 2, another 50 ohm
resistor between pin 3 and ground, and pin 1 to ground.

A balanced cable can then be used to connect both ends, putting the
resistors in the unbalanced end. It would be something like this:

50R
RCA /^^^^^^----------o 2
| o-------------o 3
GND | o 1
|50R |
| GND
GND

As you can see, both hot and cold terminals "see" a 50 ohm resistor.

What I am not sure of is how far that trick can go in rejecting hum and
noise.


Carlos

carlmart wrote:
My guess is some of you know the trick of using a "fake-balanced" cable
to connect an unbalanced output to a balanced input.

[description of 'impedance balancing' removed]

What I am not sure of is how far that trick can go in rejecting hum and
noise.

It goes as well as:

1. your estimate of the output impedance of the driving stage. If you
know the impedance precisely, you can use one resistor which is precisely
equal to the output impedance. Otherwise you need to know the output
impedance so that both legs have the same actual impedance.

2. The impedance of the output stage is constant with frequency (and this
includes frequencies well above the audio band where you still want
rejection).

3. The CMRR of the input stage.

Basically, impedance balancing works well at audio frequencies, it tends not
to work all that great in the RF region but often well enough. It does not
work as well as a transformer for the most part, and often the input stage
balancing is the weak link anyway.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."