This was deleted (accidentally), but I'll try and re-summarize what I wrote previously.
The reason why a differential input (mates with balanced output) is such a good idea has a lot to do with its noise immunity. Whiy this is so is as follows:
First, in a differential (sometimes called 'balanced') input, pin3 (v-) is subtracted from pin 2(v+). Now, the signal at pins 2 and 3 are mirror images of one another - they are 180 degrees out of phase with one another but are the same magnitude (i.e. they are 'balanced' with respect to gound). Because of this, when one is subtracted from the other, the resulting signal is twice that of the signal found on pin 2 or pin 3. Look at it this way - if pin2 = signal 'A', then pin3, which is the signed opposite of pin 2, must be ' - A'. Thus, when the subtraction is performed, you get A - - A = A + A = 2A. Thus, the output of the differential amplifier stage will be + 6 dB as compared to either pin 2 or pin 3.
Second...noise. Hum from adjacent AC mains is one of the most common ways for noise to creep into a cable - especially one that traverses a path close to extension cords or otehr AC mains. Now, we have to think a bit about the physics of induction. If we consider the wavelength of a 60, 120, 240, 300 Hz (power line and its harmonics) at the speed of light (wavelength = velocity / frequency), then we can see that the wavelength is enormous, on the order of 1E7 meters. Now, if you think about the twisted pair that are in a balanced cable in the presence of this magnetic field, for all intents and purposes, both wires occupy the same point in space relative to the wavelength. Therefore, the inducued voltage will be the SAME in both wires (those corresponding to pins 2 and 3). By 'same' we really mean the same magnitude and phase of noise will be induced into pins 2 and 3.
Now, go back to the point about the difference amplifier...when this signal is 'seen' by the difference amplifier on the XLR input, the subtraction makes the noise, which is common to both pins 2 and 3, go away. That's because the noise is the same on pins 2 and 3. So as before, let's call the induced noise on pin 2 = signal 'B'; the noise on pin 3 will also be signal 'B' (because it's the same induced noise as seen on pin 2), then we get (for noise) B - B = 0. This is known as its common mode rejection; the desired signal (riding on pins 2 and 3) however passes through intact as outlined above, free of any noise that was induced into the wire.
Three. The other big advantage to this approach is that the input and output imedances tend to be a bit lower than those of RCA inputs. This helps minimize the effect of cable capacitance, which can be very dramatic over a long run (it looks like a short across the cable at high frequencies).
If you can use a balanced send, do so. Of course, to reap the benefits, it must be mated to a differential input (as you have proposed). I am running a 990 at home and using the balanced outputs for all amplifiers (which also gives you a tone more flexibility where the amplifiers can go).
As far as cables go, my nick name says it all. It is my belief that standard good twisted pair with a decent shield and standard XLR connectors (Neutrik et al) are all that's needed to make the cable perform as it should.
I hope this helps you...
Mark
Previous Topic
Index