Why study analog circuits when you use ICs? [duplicate]

Question

Is it worth studying analog circuits when there is an IC for most analog applications?

When I need to amplify a small signal I use an op-amp. When I need to filter I design a digital FIR on a uC OR buy a filter IC.

Only thing that is important I can think of is voltage and current sources. Even then, there are voltage regulators. Oscillator circuits and AD/DA converters. To me studying analog seems like a huge waste of time because at the end of the day you do not build these circuits they come ready for you in an integrated circuit. So why would I even bother studying these circuits? I could just go straight to ICs datasheet and application notes to see the example circuit.

Answer 1

This is a little like:

• An amateur astronomer asking if they need to understand optics, when all they need to do is buy a telescope and apply it. A response here is that so long as their intent is to use the telescope as a toy and to not use it to help professional astronomers, then that's fine. But, otherwise, they need to understand the various kinds of optical error sources, which and how much of these their particular instrument has, and its impact on the images they take so that they can properly process the data and produce replicable results that can be shared.
• A .NET programmer asking if they need to understand operating systems and concurrent programming techniques when all that is taken care of by Microsoft's concurrency library tools in C# and VB.NET or by F#, more implicitly. The answer here, again, is that this is fine so long as they never really need to develop anything that's scalable. But, otherwise, they do need to understand these things.

Electronics has indeed proceeded to where much that used to be more important has become less-so, for some anyway, and where the skill levels of those practicing some nook or cranny of its art and science, can vary far far more widely than they once did. It's possible to do pretty fancy things these days with very little knowledge. Granted.

While bolting boilerplate together does get one by for a while, at some point you face an application space where, lacking the thinking tools to pursue and explore solutions, you are simply out of your depth.

I suppose you can bide your time until that happens and then use that as motivation to then learn what's needed. But that sells yourself short, and in any case, that kind of life-perspective will mean that you are always on the wrong side of the power curve of others practicing in the field. If that's all that you can expect from yourself, fine. But I'd want to suggest that you expect more than what's needed from yourself, so that you will have broad scope and depth that can be brought to bear upon each application as it arrives.

Answer 2

I think what you are getting at is that you may not need to know transistor-level design since you can just use integrated circuits (such as an op-amp).

Op-amps are analog components, so designing a circuit with an op-amp would be considered analog design.

You can use an op-amp without knowing how it works at the transistor level. But sometimes knowing how it works may allow you to use it more effectively or have a better intuitive understanding about what problems you may see or how to mitigate them.

You can possibly be a good driver without knowing much about how a car works. But if you don't know how cars work, you may have trouble fixing them, for example. I suspect that most professional drivers (truck drivers, race car drivers, etc) know a little bit about how cars work.

Also, there is a possibility that your career will take an unexpected turn and you may need to use a simple transistor-level analog circuit for some reason (cost, for example).

I don't want to read too much into your question. Maybe you are an EE student frustrated because you don't like transistor level analog design or something like that. That's fine. I didn't like control theory much in university.

But if your basic attitude is that you don't want to learn any more than the bare necessity, and that learning about how things work is a chore (rather than a pleasure) there is a possibility that engineering is not a good career for you. Again, it is understandable that there will always be gaps in a person's knowledge, and also, some areas of engineering may be more interesting to you than others. That is all fine and good.