This recent feature discusses common mode rejection ratio (CMRR) in microphone preamplifiers. In a nutshell, the signal from the microphone is balanced; any interference picked up in the cable is common mode; the preamp should amplify the balanced signal and reject the common mode signal. The better the rejection, the greater the CMRR.
However, a Audio Masterclass visitor commented that rather than thinking about the microphone having two signal outputs - non-inverted and inverted - it is better to consider that the output is 'balanced' and 'floating'. Balanced interconnection reduces the effect of interference, floating the connection rules out earth loops.
But the question remains whether a balanced signal is actually just one signal, or two?
Consider the signal flow. The output of the microphone is a transformer with two terminals. The input of the preamplifier is a transformer with two terminals. The microphone is connected to the preamp through a two-conductor (plus screen) cable.
Clearly, one current circulates around the circuit. There is definitely only one current, therefore it is one signal. No doubt about that.
Suppose however that the microphone has an electronically balanced output, which a few do, and the preamp has an electronically balanced input, which most do...
In this case, the output consists of a current flowing in the circuit from pin 2 of the mic's XLR connector to pin 2 of the preamp's input, returning through pin 1 of the preamp's input (which is earthed) back to pin 1 of the mic's XLR.
Also there is an equal and opposite signal on pin 3 of the mic's XLR flowing to pin 3 on the preamp, back through pin 1 of the preamp to pin 1 of the mic.
The strange thing is that these two methods of achieving a balanced connection come to exactly the same result - they both cancel out interference. The transformer balanced circuit is more reliable because it is 'floating', meaning that it has no reference to earth. Because the electronically balanced circuit is referenced to earth, high levels of interference could cause clipping and would not cancel out correctly.
There is another option where the mic is transformer balanced, which most are, and the preamp is electronically balanced, which most are. I'll leave this as a gray area for now.
So Paul is absolutely right to say that there is one signal. And I'm absolutely right to say that there are two. And Douglas Self, who knows more about electronic design than I ever will, is right too. He studiously solves the dilemma by talking about signal lines, signal wires or signal conductors, but at one point strays into talking about the 'in-phase signal' and 'hot and cold outputs', which plainly refer to two signals.
Anyway, take a look at this diagram, from Douglas Self's site, and tell me that there are not two signals...
However, I always believe in promoting the interests of clarity, so in future I intend to adopt the following convention...
Come on the FREE COURSE TOUR