How does one "learn electronics"?

[westfw]

Everybody over here learned calculus before they went to university.

Really?   Multi-variable, vector calc, 3d, gradiants, divergence, polar and cylindrical coordinates, differential equations, assorted series, significant complex number theory, linear algebra ?

That's pretty much what I meant by "lots of calculus" (even though some of it isn't quite "calculus.")
Since what I remember and occasionally use is more basic principles and ideas, rather than fancy techniques for figuring out the value of a capacitance given a non-constant dielectric, it's not clear that your pre-university calc (and mine) wouldn't have been "sufficient."

I'm not saying that either path is "bad", per se.  But both are somewhat "incomplete", and whether either will meet any particular person's "desired goals" is ... up in the air.
I was ... very depressed ... that most EE classes and labs I took were all about measuring and analysis, and hardly any "design a practical thing."  The logic classes were slightly better; in one, the class was supposed to collectively design a minicomputer.  In a semester.  I think the prof was slightly crazy.  (Sam Goldwasser, who subsequently went on to some degree of (pre-(flashy)web?) Internet fame (which didn't particularly change my opinion!))

As another example, there is some amount of electronics knowledge required to obtain an amateur radio operator (ham) license.  There are classes and text that teach exactly that material.  They're not a bad starting point, either  (hmm.  worth linking, even:  https://www.arrl.org/arrl-basic-electronics-course)  But you'll oscasionally meet people who will quote things like "a dipole antenna should be (468 / frequency in Mhz) feet long" and have no idea where that comes from (ie they learned "enough" and stopped (OTTH, we all "stop" somehere.))  And of course it's very radio-oriented.

[tggzzz]

Everybody over here learned calculus before they went to university.

Really?   Multi-variable, vector calc, 3d, gradiants, divergence, polar and cylindrical coordinates, differential equations, assorted series, significant complex number theory, linear algebra ?

That's pretty much what I meant by "lots of calculus" (even though some of it isn't quite "calculus.")
Since what I remember and occasionally use is more basic principles and ideas, rather than fancy techniques for figuring out the value of a capacitance given a non-constant dielectric, it's not clear that your pre-university calc (and mine) wouldn't have been "sufficient."

I'm not saying that either path is "bad", per se.  But both are somewhat "incomplete", and whether either will meet any particular person's "desired goals" is ... up in the air.
I was ... very depressed ... that most EE classes and labs I took were all about measuring and analysis, and hardly any "design a practical thing."  The logic classes were slightly better; in one, the class was supposed to collectively design a minicomputer.  In a semester.  I think the prof was slightly crazy.  (Sam Goldwasser, who subsequently went on to some degree of (pre-(flashy)web?) Internet fame (which didn't particularly change my opinion!))

As another example, there is some amount of electronics knowledge required to obtain an amateur radio operator (ham) license.  There are classes and text that teach exactly that material.  They're not a bad starting point, either  (hmm.  worth linking, even:  https://www.arrl.org/arrl-basic-electronics-course)  But you'll oscasionally meet people who will quote things like "a dipole antenna should be (468 / frequency in Mhz) feet long" and have no idea where that comes from (ie they learned "enough" and stopped (OTTH, we all "stop" somehere.))  And of course it's very radio-oriented.

Instead of constructing strawman arguments, it would be better if you included all of my two-line paragraph, viz:
"Everybody over here learned calculus before they went to university. In my case our my state school syllabus included integration and differentiation of polynomials (except 1/x) for "O-Level" exams at 15."

If you want to make a point, then giving examples of what you mean is better than SHOUTING AN AMBIGUOUS STATEMENT. You did, of course, give examples as quoted above.

Anybody that imagines "all of calculus" could be taught, let alone to 14 year olds, has somewhat inflated expectations. Pre-university calculus cannot - and should not - be sufficient for any single university course nor career.

Complex numbers, polar/cylindrical coordinates, linear algebra were "A-level" subjects. (A-levels are normally taken immediately before going to university.)

As for the balance between university theoretical and practical syllabus, I was very careful to choose a course that included both. I ignored many prospectuses that didn't give an indication that was the case, and didn't visit those faculties. Returning with friends (including one on this forum!) to an open day 40 years later, I/we were pleased to see that the tradition continued, e.g. students could use the lab equipment for their home projects.

[nfmax]

One thing you learn, when you start learning anything seriously, is that you will never stop learning. There’s always something you don’t know. Eventually, there will be something that nobody knows

[PGPG]

Hello everyone.

This thread was one of the reason I decided to register at EEVblog, but later I decided to not write here anything. But I see it is all the time active thread so I will add my 3 cents not trying to argue with anyone.

The title question can be understood in 2 ways:
1. What do you suggest me to do to learn electronics?
2. How different ways you went in your way learning electronics?

Each of people here went his own way. There are good electronic engineers who had never even touched soldering iron and can say there is no need to practice if I can calculate everything, and there are others who can think practice and only practice will make you good engineer. If we take such two individuals here they will be able argue endlessly.
I went lot of practice (before knowledge) way but I accept existence of the only theoretic approach. In my case practice was what showed me what knowledge to seek.
My first contact with electricity was when I was 4 years and 4 months old. I liked to switch Christmas tree on/off by inserting plug into socked. As socked those time have very shallow edge I managed to touch both pins with my finger when they were already in contact with AC220V.
When I was 9 my father (mechanical engineer) told me 'how transformer works' so on the 30W core I wound 7 coils of primary winding and 1 coil of secondary winding and connected AC voltmeter to secondary and inserted primary into AC220 socket and ... I didn't get expected 31V at output.
When I was 10 I build my first radio receiver (LC+diode+C+2k headphones) and it worked.
With one exception (transistor radio receiver) I have never assembled any kits (there were no such things here those time) and access to elements was restricted so each found somewhere schematic I had to adopt to elements I had so I started to calculate schematics very early. I was able to calculate working point of transistor (pnp germanium those time) in my circuit but I was not able to calculate transistor voltage gain if it was not set (limited) by external elements. But with such knowledge (being 19) I was able during one afternoon without drawing the schematic to construct (in the form of a spider on a newspaper (without any PCB)) the circuit to switch on/off my (AC220) desk lamp by clapping my hands. This is my voice in discussion about how much (or rather how little) theoretical knowledge is needed to construct something working.
During my studies, my main impression was: "Oh shit, everything I know intuitively can be calculated." Being a student I designed and made oscilloscope. It was the only way for me to have one and it has been my biggest dream for years. I was so proud of it that I have send article to electronic magazine and 1.5 year later they published it.
https://archive.org/details/Re0184OCR/Re_04_84_OCR/page/n7/mode/2up?view=theater
Electronics language is schematics so you can get everything about my oscilloscope without reading a word (you can rotate pages). Each X and Y amplifiers stage gain is set by resistor values. What is important is to have enough current everywhere to allow for full needed voltage difference in each stage. To drive X/Y deflection by not too high voltage transistors (didn't had better ones) I had to reduce anode voltage of tube and time electrons spend between deflection plates determined the bandwidth. So even analyzing the circuit bandwidth was not needed. Of course you have to know that high resistance input divider has to have also capacitors, but generally only simple, linear calculations were needed to design this.
Because of it I think that it should be named simple electronic circuit (even those time I thought of it as a complicated one).

I'd like to conclude that having practical knowledge without high theoretic knowledge allows some kind of usable electronic circuits successfully do. But there are other electronic circuits you are not able to do without deep theory understanding.

[Picuino]

One of the most elusive theoretical insights is feedback theory. It is easy to find feedback in virtually any electronic circuit and it is not easy to calculate the input or output parameters without a theoretical understanding of the subject.

As this simple circuit shows, what output impedance will it have? What gain?

You can calculate it with the transistor parameters, but it is easier to interpret it as a feedback circuit (with voltage output and current input).

https://www.falstad.com/circuit/circuitjs.html?ctz=CQAgjCCsB0AM8MYsBTAtAdhAZltM2AbABxHHElgCchhUsUIALNlOmGAFAAuIATHyYga-DH2F0GENNnx8wTFlUiQCVPuqx5C1Qk3jZI2GrEgZI4eCAAmKAGYBDAK4AbbpwBOo8WjB0+YszEIAxMGLCe3kHg6tEMAgicAOZRfvGBYBiSIZwA7qmE6eIc4hH5AeLERXF5MeJMwdTiImV1EvyC7REASlG+-oFoDSEj+owME9CQnADyUcOG9Y05Xk3tJV3M4fDJIHrghXtCmdkRAG5H7ftpIxD6DBYTIVO1+yIbLZwA9sKXobBUYKPfC3ThAA

[tggzzz]

One thing you learn, when you start learning anything seriously, is that you will never stop learning. There’s always something you don’t know.

Precisely!

And that's a very good thing too

After I've "returned" to electronics after a (pleasing) time in software, I've confirmed

• electronics learning is about fundamentals that remain more-or-less unchanged over time. There are more fundamentals yet to learn
• software learning is about the currently fashionable framework, which buttons to press, and flattie-vs-3D GUI widget fashion. Boring.

Yes, that's an exaggeration, but there's still a lot of validity there.

Eventually, there will be something that nobody knows

And the step after that is to work out which things nobody can ever know.

[tggzzz]

This thread was one of the reason I decided to register at EEVblog, but later I decided to not write here anything. But I see it is all the time active thread so I will add my 3 cents not trying to argue with anyone.
...
Each of people here went his own way. There are good electronic engineers who had never even touched soldering iron and can say there is no need to practice if I can calculate everything, and there are others who can think practice and only practice will make you good engineer. If we take such two individuals here they will be able argue endlessly.

Welcome

My main argument is with people that loudly state practical exclusive-or theoretical is The True Way. We need both practical and theoretical; both are necessary, neither is sufficient.

<many good and useful points omitted below>

I'd like to conclude that having practical knowledge without high theoretic knowledge allows some kind of usable electronic circuits successfully do. But there are other electronic circuits you are not able to do without deep theory understanding.

Precisely

[MarkT]

Study math in school, you will need a ton of it if you decide to pursue electronics in college.

Certainly upto the level of ODEs and fluency with complex numbers and Fourier and Laplace transforms.  And bits of discrete maths like coding theory, and of course computing and all things digital.

And read datasheets.  Lots of datasheets!

[wasedadoc]

And read datasheets.  Lots of datasheets!

And their Application Notes.

[tggzzz]

And read datasheets.  Lots of datasheets!

Yes, lots.

That way you can do a "compare and contrast" to work out what a datasheet isn't guaranteeing.

Reading what has been omitted is a key skill in all works of life... "I won't give you a jam sandwich before you give me £100". What happens after you give me £100?

Plus, as has been noted, read the application notes.

[Teledog]

To each their own.
I'm sure I'll get 'flamed" here, as usual.
I was always fascinated with electronics as a kid, then took a home study Heath DC electronics course with the local college, back in the 80s, it was ~$20 which included talking/interviews with an electronics instructor.
Then decided to take full time course (1 year) in digital electronics  ..bought a ZX81, modified it & all
A few years later got into ham & took another 1 year RF/coms course.
But most education was learning from "good" people around me.
Best wishes! G'Luck!
 

[Shay]

1. Read.

2. Watch the correct type of videos.

3. Exercise what you learn regularly.

4. Think.

Let me explain what I mean by each statement:

1. Read - The more you read, the more you gain knowledge, so you are able to apply this knowledge to designing projects. It's super important to read books, articles, magazines, etc. Learn from everyone, just filter the shit from the good.

2. What I mean by correct type is don't watch retarded videos that bloat you with useless, wrong knowledge. Just like you talk with people, some are OK, some are better if they are not near. So try to watch creators that have good experience in electronics, for example EEVblog or sambenyaakov.

3. Knowledge is just neurons firing in your brain. This, by itself, does not make your computer work. So you have to actually apply it in order to get practical experience. Let's say you just learned about LC circuits. Now it's time to build one on the bench and start playing with the values and making measurements. Challenge yourself - ask yourself "what would happen if I reduced the capacitance by tenfold" and try to answer the question with the knowledge you have, and then perform an experiment. This makes everything you learned in theory click together like a lego in your brain.

4. Think, be creative. Try to design things and test them, see why they don't work or work, pick up an interesting idea or a project you want to design, take your time to play with ideas related, and dont always go to the same route as everyone, maybe think about a few design choices and try to decide which one is better. Simply thinking is very important, as it raises ideas and questions with "positive feedback" that always makes you learn more and more. (E,g one question into 2 which involves into 4, etc..)