Hacker News Reader

(Ultimate Electronics Book author here.)

Thanks for the feedback. I had a similar experience starting with one of those 50-in-1 kits! And for me it didn't entirely click until I studied more of the theory and then looped back to the practical.

I've decided to mix both the theory and the practical here, and if "conventional academic textbook" is on one end of the spectrum and "Arudino" is on the other, then I'm aiming somewhere in the middle.

For a recent anecdote: this week, a coworker asked me at lunch, "How does the wall receptacle know to send more power to a 1500W space heater than to my MacBook charger?" And that's a great question. And in 15 minutes we ended up talking about fields and forces and resistive materials and the microscopic origins of Ohm's Law and hydraulic analogies... and somewhere in there, it seemed to start to click for him. And that foundation opened the door to his next question, "So why can't we hear the electrons colliding into the wire?" And that opened the door to a new topic: of course we can "hear" them if we just "listen" at the right frequencies, as analog electronic noise.

I'd love to be able to communicate that sort of intuition on a wider scale than the lunch table!

This is really good. Please keep it up.

This is a really cool project! I dig the clear explanations. Although I have a base of this knowledge already I look forward to when the more advanced topics are discussed. I'd love for a breakdown of more advanced, real-world, circuits that are broken down by subsystem and explained. cheers

I appreciate the response. I hope my story didn’t give the wrong impression. There are a lot of paths to enlightenment, and who knows what 8-year-old me could have done with an interactive book.

I could only view it through a mobile device earlier, and now that I’ve had a more detailed look, I see how the simulation and experimentation aspect adds another dimension to the theory. I intend to work my way through the whole thing and see how knowing more of the math will influence my future tinkering.

I'd draw a distinction here between electronic engineering and the art and craft of electronic design.

This is clearly an engineering textbook, starting with theory and moving on to analysis, understanding a system through mathematical modeling. If you really want to know everything (or at least most things) quantitatively about a circuit before you build it, these are the tools you need.

The parallel skill set is electronic design, sticking components and circuits together to make things. As you build bigger and more complex things, you learn more rules of thumb about different components and how to combine them into something useful.

Both of these skills together are needed in some measure in order to really be good at electronics, but everyone is more suited to one approach or the other, and learns best when they lead with that approach and fill in with the other one. For myself, I tried to learn electronics with those 50-in-1 kits but just 'didn't get it' because I didn't know what the different components did. With first year EE under my belt it all made sense and from then on I was able to learn the tinkering side.

I guess the mathematical introduction is what I was always looking for. Do you have any good beginners reference to learn Electronics Design?

user?id=taneq was contrasting "electronic design" with the mathematical approach, so not sure which one you want.

The ur-book for EE is Horowitz and Hill's The Art of Electronics. It can be found online with a little scrounging or for $100+ on amazon. H&H is extremely dense, fairly comprehensive and more like a handbook than an introduction. If you are very math-oriented maybe it will work for you, but be warned that this is over a thousand pages of formulas, greek letters, graphs, and subscripts and superscripts. It's... taxing.

As for a higher-level friendly introduction like taneq was talking about, there are a ton of resources that all cover small but essential parts. Unfortunately that means a ton of repeating things and difficulty in bringing everything together. Arduino resources are great, the adafruit/sparkfun articles/blogs are great, whitepapers from TI and others are great. I don't know of a single atlas to bring these together or say them in a single place, which sucks. I may try giving it a shot- I'm certainly math-dumb enough to understand how to translate.

The EEVblog and sparkfun youtube channels are excellent, particularly for PCB design. IMO PCB design is essential to transition from tinkering to a true hobby. Most sophisticated components only come in PCB-only packages. PCBs are far cheaper than breadboards, and mandatory for any project with more than a dozen parts. PCBs make debugging far easier. They're required for anything operating over a few MHz, and most digital stuff. Unfortunately the software still kind of blows- Kicad is the best, but still a huge pain.

I can't recommend electronics enough as a hobby! It's more intense than brewing beer, but the scene has blown up exponentially in the past two decades and is incredibly accessible. Electronics are more affordable than any other engineering discipline- PCBs are simple and incredibly cheap to order in single lots, and 80% of components can be ordered in single units. Compare eg metal prices, which are easily 20% the price in bulk vs. small units. Single electronic components are 75-50% of the cost in bulk. Entire industries are dedicated to making cheap, simple modules that handle incredibly sophisticated tasks like location tracking, video, wireless communication, or battery power. You can do anything you can think of.

Ha ha, "If you are very math-oriented maybe it will work for you". The AOE is definitely not math oriented, and not what is typically used for EE courses. If AOE was used for undergrad EE, then 2/3 wouldn't flunk out. Don't get me wrong, it's a good book, but it's definitely not for the math oriented.

Lots of great points, but perhaps a bit unfair to AoE. As madengr says, the math gets way worse than what's in AoE.

AoE isn't really supposed to be a textbook in itself. It was written for physics students, typically at the graduate level, who need to design experimental apparatus without a formal engineering background. It's not an ideal introduction for newbies, and unfortunately it's recommended for that role way too often IMO. But it's a great sophomore resource, so to speak.

What's needed is something between the Forrest Mims "cookbook" level and AoE... something that gives you the theoretical underpinnings needed to know what chapter in AoE to turn to. People who are interested in the RF and communicstions side have always had the ARRL Handbook as a resource, but that book has limited appeal to those who are more interested in microcontrollers and other electronics topics. This page looks like it might make a useful contribution there.

That's exactly what I mean by math-oriented- it's good for people who understand things in terms of equations. If that's all you need to be comfortable, it's great and you can very quickly find what you want to know. That's what makes it good as a reference handbook.

Maybe this is a problem of semantics, or something that I simply can't quite understand any longer because I am an electrical engineer (or at least I have a diploma that says I am :-) ). But this distinction is very strange, and I'm not sure why/how you're making it. Electronic engineering is nothing like that.

Knowing rules of thumbs and how to "combine components" into something useful is not a parallel skill set to electronic engineering, it's a part of electronic engineering. Rules of thumb, models and reference schematics (from manufacturers or whatever) are pretty much how you come up with the first draft of a circuit.

Also, electronic engineering consists precisely of designing circuits. Quantitative analysis is a big part of it because without it you can't always know whether the circuit you've designed works according to specs (and in practice, yes, experimentation and testing plays a big role, but there are only so many prototypes that you can blow up before you run out of time, and only so many things that you can test for in a regular lab). The whole point of one's activity as an engineer is to combine components into something useful.

Sure, you don't sit in a lab drawing schematics and building things all day, because there are a lot of other activities that go into building a good product. There's a lot of validation work and a lot of analysis and a lot of planning. Some engineers focus their time only on one of these things, especially because they really are so complex, and so complicated, that you can spend a lifetime studying just one of them and you're still left with a lot of stuff to know.

But at the end of the day, designing electronic gizmos is pretty much what you do, regardless of what role you're playing there.

You can certainly make things by sticking components and circuits together without a thorough understanding how they work. But the idea that engineering somehow mostly about understanding a system through mathematical modelling before building it, and design is mostly about making things by sticking components together by intuition, is very much absurd.

Good summary. I'd also add that 80% of the difficulty is spent on optimizing the last 20% of performance (i.e. the last dB), which is what separates the hobbyist from the pros, or the art from the quantitative.

In my time this was basically a distinction between an engineer and a technician - technicians generally having a lot more practical experience, but lacking in the theory, while fresh out of Uni engineers knowing the theory well, but lacking practical skills in solving the real-world problems. With time a good engineer picks up those skills too.

I found All About Circuits to be a great resource when I was a teenager. after skimming this site, I'd say it's too focused on math fundamentals, and not enough on "yeah but how does it ACTUALLY work?". I feel like All About Circuits does a good job of skimming over details that are not going to be important until you've built up some more knowledge.

Anyway, it's a great resource, and was a big part of why I decided to do EE instead of CS. It's also got pictures; lots of pictures!

https://www.allaboutcircuits.com/textbook/

and for the specific example you mentioned, their section on the bipolar junction transistor is simple and accessible:

https://www.allaboutcircuits.com/textbook/semiconductors/chp...

From the intro:

> The motto of this book is “one level deeper.”

Really we need both. Lots of people do learn by experimentation. There are (now) a lot of tutorials for them. But on the other hand there are people who really aren't satisfied until they've pinned down all the details - what is an electron really etc - and this is more suitable for them.

The more the merrier. We just need a bit of signposting too.

This mirrors my experience even beyond the hobby stage. I was quite a bit luckier because I had access to the internet and various dev boards as a kid and could work with them as a low level programmer but trying to move into schematic and PCB design was downright impossible. I couldn't wrap my head around any of the online resources and like you stuck to programming.

Fast forward to my 20s when I could afford to take a short sabbatical to apprentice for a practicing electrical engineer (self-taught as well). Within four months I went from basic breadboarding and soldering skills to designing high speed digital PCBs from start to finish and began contracting as an EE immediately. There was a little math but mostly just a bunch of specialized calculators for impedance matching high speed traces, capacitance, and length matching high speed buses.

I went back to software for the pay but when I designed a small control board last year, the whole exercise from inception to sending off to fab and assembly took four days for an 8 layer PCB. Between all the online parts databases (with footprints and schematic components!), reference and open source designs, and software like Altium and TopoR, electrical engineering has never been easier and involved so little actual theory taught in schools.

It hurts my heart that I haven't yet found a resource that just dumps people into the deep end with a proper focus on the engineering instead of the theory.

Can I ask how did you found someone to apprentice for and what did you do?

I think its exactly what I need, I feel really lost on where to go with respect to electronics learning, and I think a mentor could help me a lot.

The big breakthrough for me was the realization that a transistor is similar to a 4 pin relay, but with a shared ground between the main circuit and the control, resulting in 3 pins.

The next step is to realize that solid-state devices usually control current instead of voltage, but that we can add voltages, currents and resistances to the circuit that let us work with voltage in a linear region of a device's response curve. So our normalized 0-1 control corresponds linearly to 0-1 on the main circuit. From there, it's straightforward to build amplifiers where the control voltage or current is thousands or even millions of times smaller than the main.

After that it gets.. complicated. It took me 4 years of math and physics to finally understand solid state theory and be able to analyze large circuits by subdividing them into simpler linear sub-circuits for my degree. Then the really interesting stuff happens when we abandon all of that and convert to the frequency domain using the Fourier, Laplace or Z transform. So discrete signals have periodic frequencies and periodic signals have discrete frequencies. Which lets us analyze the transient and steady state portions of a signal separately and gain valuable insights about what a circuit will do.

Of course I've mostly forgotten all of that. One word of advice: keep your college textbooks. I still find concise explanations there that haven't made it onto the internet over 20 years later.

Edit: IMHO, the above way of translating between abstraction and application is the primary advantage of getting a degree from a university. For anyone young reading this, I highly recommend applying to the best schools you can. If you just settle, then you might miss out on the underlying theory behind each discipline. You'll want the theory later when you've forgotten everything like everyone else, because you'll be able to re-derive everything you've learned from first principles when you need it.

I think for me some of the difficulty in understanding electronics comes from the way components are often used in ways that don't make any sense unless you happen to know the 'trick' involved. For example, using a reverse biased diode as a constant voltage reference, using an inductor to create a high voltage pulse etc.

Also, for what it's worth, "Practical Electronics for Inventors" is one of my favourite books on electronics. It's well balanced between theory and practical applications.

I guess there are at least two ways to understand an inductor:

1. if you disconnect it from power, you'll get a huge voltage spike 2. V = -L (d/dt) I

1. seems like a practical explanation. 2. seems like a theoretical explanation. But an expert can play with both, and (under the assumption of first-order linear ODEs for this primitive components, which is an assumption you can only make after the fact, so... bare with me for the following BS statement), they are both equivalent. (That is, I can't think of another relationship that achieves 1.

Your idea of a "trick" has mostly to do with 1, I think. And it's an essential part of understanding an inductor, and it's also most likely how humans discovered inductors.

Sure but neither of those explanations immediately scream you can use this to filter high frequency noise or you can use it to cause a phase shift in an AC signal (like in a mechanical electrical meter or a start winding in a single phase induction motor).

Maybe 'trick' is the wrong way of looking at it, I guess the real issue is that from seeing a single component in a circuit diagram you can't know which of the effects of a component is being utilised.

I think I better understand your point. There's behavior, and there's intent. Seems related to the distinction between syntax and semantics.

The way I learned how an inductor works: it doesn't like changing current. When you turn the power on, it sluggishly and reluctantly lets the current flow. Then if you try to stop the current flow, it uses the energy it grabbed earlier to crank up the voltage in an attempt to keep the current flowing.

An NPN transistor, when saturated, is like a SPST relay where one side of the coil and one of the contacts are internally connected to ground. You don't however end up with isolation between the input signal and output signal in either case. Of course, the way a transistor actually works is dramatically different than a relay - there are plenty of useful things to do with a transistor when it's not saturated (plus it's a lot easier to blow the transistor up).

I wouldn't invoke a notion of "ground" when describing the operation of a component. That's some really a notion that comes out of looking at a circuit (in the sense of a circuit diagram), I would say. It might also confuse a beginner.

Otherwise I think your explanation is great. I can't claim to understand transistors super well, but your explanation touches on a key point for someone who is trying to understand the relationship between relays and transistors. Transistors cannot achieve the impossible--with 4 pins, you can achieve electrical isolation between the switcher and the switchee (and share a "ground" if you want to). A transistor cannot do that.

This might also be a high-level way to see why we need two "kinds" of transistors. Because suppose the electromagnet in the relay has some polarity, and only activates if current flows "north to south". Well, there are two possible choices of what side of the coil you tie to a pin on the switch, and what side of the coil you expose as the base/gate.

True ... ground in this instance could be a confusing word. I guess choosing an NPN transistor led me to choose that word but it's really more about the emitter being negative with respect to both the base and collector (kind of wordy). And the transistor becomes even more interesting because if the emitter becomes more positive than the other two connections, the base-emitter junction becomes a reverse-biased diode and blocks current flow completely.

I went through the same progression as you did. Dave Jones' OpAmp video was a godsend, I wish he'd make more videos like that. I think there's a great deal of value in "slow learning" when you have the time; let complex ideas percolate through your subconscious through repeated exposure over time.

It's been said many times before but I'd like to recommend The Art of Electronics to anyone who's graduated from the YouTube School of Electrical engineering and has some motivation to keep going. It has some mathematical theory (which it insists is optional) but I've found it useful for filling in the holes in my education and has loads of practical advice like the EE mentor I've never had.

Or just buy The Art of Electronics :)

Great book, but IMO too advanced for a beginner. The risk of being discouraged by the level of its topics is high. I would rather suggest to a complete newbie to start experimenting with analog electronics by following well known old magazines, soldering kits, then attempt to modifying them, destroy components by applying the wrong voltage/current/polarity (or sometimes just by touching them with ungrounded hands) and learn why and how to avoid in the future; in other words learn by failure and success. I would also initially stay away from Arduinos, Raspberries etc. They're great but they also hide the inner workings of a circuit behind the code (and often also the code behind a linkable library). Building a single transistor wah guitar pedal for example will teach a lot more about electronics (and some mechanics too) than any small computer module out there: they will be wonderful addition to make great things with the knowledge gained through simpler projects.

So you want to be a plumber? Here take this graduate course in fluid dynamics.

I’m an electronics tech and i feel similarly about music, my hobby is to make electronic music but whenever i wanted to grow my compositions i encountered lots of theory and maths but i realized that without the theory i was limiting myself to just follow instructions. So if you want creative freedom in whatever you do from electronics to baking you need to understand why you’re doing what you are doing.

Bret Victor is really big on this. Many people (including myself) learn a lot by experimenting and learning the dynamics of some system. One just needs a system to play with, control and feedback.

I was the other way I wanted to know how it all worked in great detail.

Like magnets and EMF the forces is explained by showing lines of force but what are they? From what I understand the force is energy due to the exchange of virtual photons between the magnetic poles.

I've also read that EMF is also due to relativistic effects. Moving and stationary charges differ in quantity due to contraction of the moving charges. Like charges repel so the difference creates a force.

Take all that with a huge grain of salt.

That's right. Here is a pretty good video explaining magnetism as a relativistic effect of electrostatics - https://www.youtube.com/watch?v=1TKSfAkWWN0

The Radio Shack 50-in-1 electronics kit was also my onramp to technology, and I too landed in programming rather than, e.g. electrical engineering. I recognize the loss you describe... and I'm really grateful, all the same, for the door it opened into the future. Glad to see you found an alternate on-ramp.

Avoiding “equations” when you want to understand theory is a mistake IMO in the same category as avoiding “code” when you want to understand software: it’s possible but it’s definitely the harder way to do it.

Not joking or being sarcastic here, Electronic Circuits for Dummies is great for what you describe.

Are you my twin brother from a parallel universe? My experience has mirrored yours.