"Explain Like I'm 5": Opamps July 21, 2015 14:09
What are opamps?
Opamps (or op-amps, or operational amplifiers) are small, inexpensive integrated circuits that can be used to do a ton of different things in audio electronics. They can apply gain or attenuate a signal, create filters, present desired input/output impedances, oscillate at specific frequencies, etc.
They’re usually manufactured on a black, spider-like component with at least five terminals:
What do opamps want?
Instead of talking about how opamps work from an electronics perspective, it’s easier to think about what they want. Opamps want their inputs to be equal all the time. So if the voltage at both inputs is 0, the opamp is happy–it doesn’t need to do anything at all. But the second we change the voltage at one of its inputs, the opamp will jump into action immediately to give the other input that same voltage. Absolute equality–that’s all it wants.
Opamps are so committed to this single desire that they will burn themselves out in a plume of noxious smoke before admitting defeat.
How do they get what they want?
Opamps use their outputs to make their inputs equal. But they can’t do this on their own–they need us to provide a feedback path. Let’s say we kindly oblige and solder a wire between the output and - input pins. (This is called negative feedback.) Now whatever voltage the opamp sends to it’s output gets immediately sent to the - input as well. In other words, if the + input is fed a certain voltage, the opamp can immediately make the - input the same by adjusting its output voltage. In other words, it can get what it wants!
Now let’s do that with numbers so you can see what it looks like. Say we connect a microphone with a 1V output to the opamp’s + input. The opamp wants to make - input 1V as well so it sets its output to 1V. And since we’ve attached a negative feedback wire, that 1V is immediately sent to the - input. Now both inputs are sitting at 1V and the opamp is happy.
How we make them do what we want?
That’s all well and good, but if we stopped there the opamp would only be good for passing unity gain. The real fun comes in replacing that feedback wire with some more interesting components. Let’s say we replace it with two resistors configured as a voltage divider.
A voltage divider in the negative feedback loop
These resistors will take the voltage from the output and cut it in half before it reaches the - input. Let’s go back to our example from the previous section. If the output were still set to 1V, the - input would be at only 0.5V. And that’s not what the opamp wants! So now it adjusts its output to 2V to get 1V back to the - input. The opamp is happy again and our microphone signal is 2x (6dB) louder at the output of the opamp.
And gain is just the beginning. We can put all sorts of stuff in the feedback loop: capacitors and inductors for filters, diodes for clipping, transistors for variable gain–you can even put other opamps in there! The point is that the opamp just wants one simple thing–to make the inputs equal–and we can make it do all sorts of stuff just by making it work harder to make that happen.
How do opamps work?
I honestly have no idea. I’ve never designed one or even bothered to look at one’s schematic diagram. But that’s the beauty of integrated circuits: you can treat them like a black box. If you understand their specs and theory of operation, you can use them without knowing what’s going on inside.
Why are there so many kinds of opamps?
All of design is basically managing tradeoffs: more gain vs. more noise, greater bandwidth vs. less stability, greater precision vs. higher cost, etc. There are hundreds of different opamps and they all do the same thing but make different tradeoffs. So one opamp may be ultra-low noise, but have stability issues at high gains. Another may be very low cost but have lots of noise. And so on.
What are discrete opamps?
99% of opamps you’ll come across are monolithic, as opposed to discrete. They look like the spider thing below and are manufactured in massive quantities on silicon chips.
A standard monolithic opamp IC
But some folks in the audio community are not content to pick from the variety of monolithic opamps available. They prefer to roll their own by placing individual components on a printed circuit board. They are big and expensive and usually sport “worse” specs than even low-end monolithics, but they have certain benefits like being proprietary and high-margin entirely customizable to suit the designer’s ears.
Image courtesty of 6SN7, licensed under a Creative Commons License
Do different opamps sound different?
There's much debate over how significant the audible differences are between different opamps. Some people (often musicians) claim to hear a "night and day" difference when they swap the opamps in an old piece of gear. Some people (often engineers) claim there's no difference at all.
I've spent a good bit of time listening to different opamps in a controlled environment and my take is that audible differences between opamps are vanishingly small when they're operated in their linear range. When you start to ask them to do stuff beyond what on their spec sheets, all bets are off.
So, swapping an opamp that's just a buffer between two stages in a compressor? Probably not gonna hear much of a difference. Swapping opamps in a circuit with a hot input signal and excessive amounts of gain? You will hear major differences, akin to using two different mic preamps to apply a lot of gain.
Did I miss anything?
I hope that helped clarify things a bit! Is anything unclear? Let me know in the comments if you have any other questions about opamps I can answer.