Common mode rejection

In yesterday's post we saw how easy it is to design with an IC op amp. All you need are two resistors and two 9 volt batteries and you've got yourself a good sounding line stage. Add a pot (volume control) and now you've got an entire volume controlled preamplifier that runs on batteries. Add some fancy connectors, a nice looking box and you too can join the ranks of the high-fi audio manufacturers. But we also discovered that different IC op amps sound quite different from each other as well we learned something very fundamental about op amps as a category. Op amps are not necessarily IC's. If you get nothing out of this series than that piece of information, grab it and be thankful you get it and understand. An IC can have many purposes; one of which is that of an op amp. An op amp is a functional amplifier block. It can be made with vacuum tubes, discrete transistors, or on a single piece of silicon. An op amp is not necessarily an IC so now when you hear the term used to describe the insides of a piece of high-end gear, you can ask the right question. "Does your design contain any IC amplification devices in the audio chain?" Don't ask if it uses op amps because op amps are not IC's they are very useful building blocks that, designed properly, are wonderful sounding and extremely useful. You just need to know if the audio circuits designers used are integrated or discrete. As we move through this series I'll explain some of the good and bad elements of integrated circuit amplification devices so you'll understand also what it means. But now, let's get back to op amps - remembering they are a functional amplifying block - meaning they are a predesigned block that can be used in a multitude of ways. The macro functional block we call an op amp is made from three micro blocks: a differential stage, a gain stage, a output stage. Here is a diagram of what it looks like inside. Today we are going to focus on the first micro block, the differential stage. Notice in our little diagram of the op amp that the two inputs we learned about yesterday, the - input and the + input are both connected to one of the two inputs on this differential stage. What is the purpose of this stage? It is our input to the op amp and its function is really cool. It amplifies differences between the two inputs just like its name implies. Why is this cool? Well, there are several reasons but mostly because of what it doesn't do. It doesn't amplify common signals. This is something important to understand so in today's post we're going to focus on this all important aspect of our op amp functional block. If you put the same signal into both inputs on our differential stage then the output will be zero. Nothing, nada. If you put something into one side and not the other, then the differential stage will amplify and give you a bigger signal than what you put in. Depending on which input you put the signal into determines what you get out. Place a 1 volt sine wave into the + input and you get an identical looking bigger sine wave on the output of the differential stage. Place a 1 volt sine wave on the - input and you get a reverse order bigger sine wave on the output of the differential stage. We refer to this reverse order sine wave as inverted (upside down) and this is what happens when you use a phase inversion switch on your preamp or DAC - the signal gets inverted. Designers can simply choose which of these two inputs to feed the audio signal into with the switch and you get either phase correct or phase inverted out (which is another cool feature of this stage). Place an identical 1 volt sine wave on each input and what you get out is zero. OK, why is this important? Because if used properly you can reject signals you don't want while simultaneously amplifying signals you do want. This process is called Common Mode Rejection Ratio or CMRR and it is used to eliminate noise and hum which can be common to a signal. CMRR is most often taken advantage of in XLR balanced cables but it can also be used in single ended RCA cables as well. How does this work? Imagine a two-wire cable going from a turntable to a preamplifier. On one end you have the phono cartridge which is a coil of wire - with a beginning and end wire - each end of the coil is connected to one of the two wires of the cable. The other end on the preamp has a differential stage amplifier and on each of the two inputs we place the other end of the wire. With me? Now we start to play a record. The coil generates a moving voltage which is different at each end of the coil - it's AC so it's going + to - and then - to + so the voltage is moving back and forth over the coil - the signal always the opposite on each end of the coil wires. Our differential stage is loving this - it amplifies the differences between each end of the coil as the signal moves back and forth and we hear music. Now imagine a noise source - hum from a nearby transformer, noise from a cell phone or anything radiated in the air. That radiated noise is going to pass right through our two wires and be present on each of the two wires in equal amounts. So each of our two wires has lots of noise on it but the noise on each of the two wires is identical. The noise is common to the two wires. What does our difference amplifier do with this? Nothing! It rejects the noise completely. So you have high noise and signal coming into the differential stage and only signal coming out with the noise gone! It's a bloody miracle. And there, my good readers, you now understand Common Mode Rejection, how it is rejected while at the same time amplifying only the music. XLR cables do this best because they have the two wires inside them surrounded by a third wire that is called a shield. An RCA cable has only one wire and the shield and they are not equal in construction so the noise isn't as common as it could be and thus rejected not quite as much. Tomorrow we find out how this works.