Balanced and Unbalanced

[mbenonis]

I feel like there is a lot of confusion as to what it means to have a "balanced" or an "unbalanced" interconnect. These two articles are by far the best explanation I've ever read:

http://www.jensentransformers.com/an/an002.pdf
http://www.jensentransformers.com/an/an003.pdf

Both articles come from Bill Whitlock at Jensen Transformers, and are accessible here:
JENSEN TRANSFORMERS, INC. - APPLICATION PAPERS AND SCHEMATICS

Please feel free to post any questions you have after reading these articles and I'll do my best to clarify.

[museav]

The concept that impedance balance, signal symmetry and differential inputs that are all commonly associated with "balanced" signals were actually three separate aspects was a revelation for me and the direct result of a Bill Whitlock presentation, of which I 've had the privilege of attending several. Understanding this has made understanding many other related issues much easier.

[Andy_Leviss]

I would slightly modify that to suggest that impedance balance and differential inputs go hand-in-hand; without one, the other is pointless. Signal symmetry, however, is indeed quite separate, and inconsequential for the purposes of noise-reduction.*

Thanks for posting this, Mike!

--Andy

*-No matter what Frank Wood might believe, for those on the Stagecraft List!

[museav]

For me the simple realization was that any noise reduction aspect had to work in the absence of any audio signal or it was sort of pointless, therefore the noise cancellation had to be independent of the signal. Thus signal symmetry is not relevant to the noise cancelling aspect of a balanced line.

A differential ciruit looks at the difference between the signals on the two conductors, so any signal that is equally present in the two conductors is then cancelled while any noise that is differential between the two conductors becomes part of the audio signal. Impedance balancing helps insure that the noise introduced onto each conductor in a balanced pair is the same, thus maximizing the effectiveness of the noise cancellation from a differential input. So two separate aspects but they do indeed go hand-in-hand.