Why I Hate Filters

[syntonica]

They are not logical little critters that do as they're told! If they do, they're uninteresting and if they don't, well, all sorts of blow-ups happen!

It's not the math that gets me, it's how the math represents physical circuits. I can do the math and work it out on a theoretical level and I can read a diagram well enough to build it, but my understanding of circuits is so poor, I'd be hard pressed to even tell you which way the little electrons are running, even while staring at the transistors and diodes. I'm more 1s vs 0s than continuous voltage.

I need to expand my knowledge of circuitry to where I can model them in Spice and basically understand what each component is doing and why.

So, my request is, is there a book/tutorial I can use to help me? I'd like it to include using a Spice-like application. It doesn't need to be ultra-basic 101, but it can't be overly rocket science complicated. I guess I need a Goldilocks solution here.

Once I get what the math is doing, I can herd my filters rather than try to strong arm them.

[Fire Sledge - Ohm Force]

Some classic books:
– For electronics in general: Paul Horowitz, Winfield Hill, The Art of Electronics
– More specifically for filter design: Don Lancaster, Active-Filter Cookbook

[syntonica]

Some classic books:
– For electronics in general: Paul Horowitz, Winfield Hill, The Art of Electronics
– More specifically for filter design: Don Lancaster, Active-Filter Cookbook

Thanks! Of course, since they are probably chock full of practical information, they are expensive!

[Ivan_C]

This one is free:

https://link.springer.com/book/10.1007% ... 19-21173-2

[syntonica]

This one is free:

https://link.springer.com/book/10.1007% ... 19-21173-2

Thanks! That one looks to be quite useful. It's going to be slow going with relearning the math. Vector calculus, I studied some 40 years ago!

[mystran]

This has a lot of useful stuff that will help guide you http://qucs.sourceforge.net/tech/ when it comes to turning circuits into simulation. The electrons can go both ways anyway, so there's no need to worry about that too much.

Specifically you are interested in "transient analysis" but the other stuff is potentially useful to learn as well.

[syntonica]

This has a lot of useful stuff that will help guide you http://qucs.sourceforge.net/tech/ when it comes to turning circuits into simulation. The electrons can go both ways anyway, so there's no need to worry about that too much.

Specifically you are interested in "transient analysis" but the other stuff is potentially useful to learn as well.

Thanks! That will be really useful once I get more of the basics down. Unfortunately, I haven't studied electronics fully since 7th grade, and that was only to get access to their new-fangled TRS-80s!

[mystran]

This has a lot of useful stuff that will help guide you http://qucs.sourceforge.net/tech/ when it comes to turning circuits into simulation. The electrons can go both ways anyway, so there's no need to worry about that too much.

Specifically you are interested in "transient analysis" but the other stuff is potentially useful to learn as well.

Thanks! That will be really useful once I get more of the basics down. Unfortunately, I haven't studied electronics fully since 7th grade, and that was only to get access to their new-fangled TRS-80s!

I used to have a lot of trouble understanding how electronics works, because in my experience a lot of electronics literature involves a lot of vague hand-waving that just makes things more confusing. It wasn't until I started learning about circuit simulation where the underlying math really has to be made explicit that the whole thing finally started making sense.

YMMV.

[syntonica]

I used to have a lot of trouble understanding how electronics works, because in my experience a lot of electronics literature involves a lot of vague hand-waving that just makes things more confusing. It wasn't until I started learning about circuit simulation where the underlying math really has to be made explicit that the whole thing finally started making sense.

YMMV.

I tried the first page. I was like, WTF are you talking about? It would make more sense to me if it was in Mongolian!

I still need to get many more basic electronics concepts under my belt.

[DaveClark]

Hi syntonica,

In your original post here, you mentioned several different subjects in electrical engineering and physics. You mentioned filters, but also electrons in devices. There are many different levels of both math and physics involved in the physics of devices. Some math involves only terminal behavior while other math, combined with physics, is applied to tiny details of behavior, understanding of which requires a knowledge of (for transistors anyway) solid state physics which itself requires some understanding of quantum mechanics. With respect to inductors and capacitors, the dirty details would include knowledge of electricity and magnetism whereas terminal behavior is fairly simple by comparison. With regards to filters, you mentioned 1s and 0s but also referred to circuit elements. As you probably know, there are different types of filters, for example analog and digital, or continuous and discrete. Because you mentioned active devices (transistors and diodes), you should know that there are also passive and active filters. You also mentioned SPICE. Doing SPICE simulations is different than writing code for approximating (for example) passive filter behavior for real-time modelling.

It would be very difficult to suggest a single reference that would cover all of that without being a very thick handbook type of thing, and probably not very useful for you at this time.

So --- unless you can provide a clearer picture of what it is you really intend to do, it would be difficult to suggest anything other than a pile of books for you to study over the next two years or so (see below!).


Now there is a book called The SPICE Book by Vladimirescu which is about using SPICE, but has possibly enough discussion about circuit elements to give you what you are looking for (i.e. terminal behavior and a bit more). It’s not about filters, however, which is the subject of this thread according to your subject line. But I got the impression that you are determined to learn SPICE and to understand at least a little bit about how circuit elements work. For further study of transistors and diodes, there is a book called Semiconductor Device Modeling with SPICE by Massobrio and Antognetti, but that’s probably more than you want to know about SPICE models, being a pretty thorough discussion of transistor and diode modelling, albeit now dated. Even further along, there are books on device physics such as the classic by Simon Sze and The Physics of Semiconductors and Their Heterostructures by Singh.

If your objective is actually to write code for emulating RLC filters in real time, then you probably should look for a good book on analog filter design such as Analog Filter and Circuit Design Handbook by Williams (affordable and quite complete) or a book about both continuous and discrete signal processing such as Continuous and Discrete Time Signals and Systems by Mandal and Asif. If instead your interest is solely in digital filters and writing code for them, then you should probably look for a good book on digital signal processing such as Oppenheim’s and Schafer’s classic book Discrete-Time Signal Processing, or its more recent derivatives. There actually is some very practical discussion of digital filter design contained in a couple of books by Will Pirkle, namely Designing Audio Effect Plug-ins in C++ (first and second editions) and Designing Software Synthesizer Plug-ins in C++. However, these books contain far more than just filter design. Then, of course, for general study there are the introductory books for EEs in 1) Electronics and 2) Circuit Analysis. The former is a more math-oriented version of books that other posters have already recommended about the various kinds of circuits and what they do while the latter is involved with calculating currents and voltages, producing results like those from SPICE, kind of a “before SPICE existed,” i.e. matrix algebra, Laplace transforms, etc.

Again, if you could be more specific about what it is that you plan to accomplish and the path you expect to follow, you may receive more appropriate suggestions.

Apologies to all for this long response, but syntopica’s opening post invited one.

Regards,
Dave Clark

[syntonica]

Hi Dave, thanks for the lengthy reply! Sorry if my request was rather vague and confusing.

Basically, I see three steps on what I want to learn:
1. Basic electronics: what components do. I want to be able to look at a circuit diagram and recognize a filter or a BBD and be able to follow what's going on. I don't necessarily need to be able to design anything.
2. Spice: I'd like to be able to model a circuit in a Spice-like program to study it further and see what each component is doing.
3. The Math: I'd like to be able to translate 1. and 2. into the corresponding formulas and to be able to solve them. I have problems understanding math unless I can understand what the math is saying about some physical reality. There's a reason why I struggled with theoretical physics.

I single out the filters as that is the one thing I just can't figure out. Start talking about poles and zeroes and I'm clueless. I can do the math, but it's meaningless to me without understanding what it's modeling. If there's a book that combines 1 and 2, that would be fantastic!

You've recommended a number of books. I'll have to check them all out to see if I click with any of them.

[DaveClark]

Hi syntonica,

It looks to me like you may want to start with an introductory book on circuit analysis as well as look at a book like the one another poster recommended: The Art of Electronics by Horowitz. Note the word "Art." Treat the book like an art gallery, not like a textbook. There is a lot of stuff in that book that you will never need, e.g. a discussion of current mirrors. Learning to recognize circuits will take time. Also, everyone has their own way of drawing schematics, so the any given circuit can look completely different from what you get used to seeing.

There is a place called the OpenLibrary where you can check out books, including hundreds if not thousands of textbooks, online for free. You have to create an account, then install a particular version of Adobe reader called Adobe Digital Edition or something such as that. They have a number of books on circuit analysis:

https://openlibrary.org/subjects/electr ... t_analysis

Note, however, that many listed are not actually in their library. From what actually is there in the library, you can at least get a feeling of what you might want or not want pretty quickly.

To start you out, there is a book at OpenLibrary by Lawrence P. Huelsman, Basic Circuit Theory (1984) that was used for introductory EE classes. If that's too terse, then perhaps Introductory Circuit Analysis by Robert L. Boylestad (1990) would suffice as a gentler introduction.

After a book on circuit analysis where you do analyses by hand, you can start using SPICE with The SPICE Book. Books on circuit analysis have, over the years, included more and more on SPICE, but you still should learn to analyze simpler circuits by hand in order to understand what is going on.

Regards,
Dave Clark

[JCJR]

Folks learn differently (duh). mystran's "light coming on" when he finally grokked the simulation math wouldn't have ever made the light glow for me.

I do not know if I was younger and tried to learn a little electronics via SPICE simulations, simulated signals and simulated scope traces, whether that would light my bulb. Maybe so.

The hardware equivalent finally lit my bulb at least a little bit many decades ago. Old Dad was an EE and his friends were EE's. As a kid I would solder up stuff and read electronics texts, but such as bias, feedback and gain really never connected enough to make practical sense. I'd ask questions of Dad and his buds and they would answer what they thought was clearly, which sounded like gobbledegook.

Then old Dad gave me an oscilloscope, and I already had some other test gear, and bought a breadboard and built a signal generator and power supply and started making test circuits on the breadboard and poking around on the breadboard circuits with the scope. Seeing how different points in a circuit behave, gradually figuring out why they were behaving thataway. And with the "intuition" from observation, the math of things such as bias, feedback or gain began to make more sense.

In theory you could do exactly the same learning lab experience with a SPICE program and you wouldn't have to sniff solder fumes or occasionally burn up components. Maybe the simulated SPICE lab would be as good or better. I don't know if SPICE would have worked on my thick noggin had I been born 40 years later. Maybe. Who knows? But it is fun and educational to occasionally burn up a circuit, and such lessons tend to "stick".

Hands on learner. Same with programming. Learned whatever programming I know, well took a few courses in college and read some books of course but most of it was practice writing terrible programs then fixing them. Maybe somebody could learn to program just by reading about it, but that would not include me.

Different stuff lights different folks' bulbs. My electronics bulb would have never got lit if I never had an oscilloscope. But you have to use it. Just buying one and putting it on the shelf is no portal to understanding.

[syntonica]

I'm on the autism spectrum, so my learning style is, well, definitely not normal and I always have learned differently. Most of the texts I read, I can follow up until they throw something in without definition, telling me I should have learned that one thing previously, but I'm never quite sure where I should have learned it.

I downloaded a couple dozen PDF textbooks on electronics and DSP and everything in between I could find. (Thank you, universities and other institutes of higher learning!) I was looking at one and they talked about an RC filter and they even stated that that was resistor-capacitor! And I was all Hallelujah! Thank you! Nobody could have mentioned it before in the twenty papers, articles and texts I read before that?!? Many specifically on what filters are!

I actually found an older text on analog filters that I've been able to follow with just looking up a few bits of mathematical notation I've never seen before. It focuses more on the practical aspects and mostly, purposely avoids the overly complicated math, so I'm starting to accumulate some of this information. Slowly... I like the PDFs since I can annotate them just as I would if I had a dead tree edition and I can carry around all of these texts without the necessary weight lifting and conditioning exercises.

Thanks to everybody who's replied with suggestions! Hopefully, they'll be of use to others who may follow, too.

[DaveClark]

Hi JCJR,

My own education and training in electronics includes practical experience as an elecronics technician, and this saved me from a serious delay in grad school. So I very much sympathize with what you are saying. But syntonica's goals specifically include SPICE modelling and deriving equations, so would still require study of circuit theory and of different circuit types, regardless of practical experience.

Regards,
Dave Clark