How to learn more electronics
November 26, 2006 9:27 AM Subscribe
I am pretty good with electronics. How could I get better without going back to school?
I am pretty good with electronics. I have the basic circuits mastered, and I know much more about the underlying physics than I will ever use. I can even work with 120 volts without electrocuting myself (so far, knock on wood). My last project consisted in salvaging components from dead computer speakers to reassemble them into a diminutive amplifier and a power supply. It's simple really. It's a matter of following the reference circuits for the voltage regulator and amplifier chip.
I wonder where to go next, how to get better. I would like to become good with electronics, perhaps good enough to assemble an MP3 player and understand what I'm doing, good enough to build my own ergonomic keyboard, good enough to control robotic step motors from a computer program. I want to be able to build large amplifiers circuits, tweak them and debug them.
The normal path is to get an engineering degree in electronics. But I already have a master in computer science, and I am 29. I don't relish spending another four years in school. I have found the lectures on electronics in MIT open content, and I have found the online version of "fundamentals of electrical engineering and electronics". They seem to focus on high-level understanding, which I already have, and not on the practicalities of building cool gadgets (perhaps that assessment is mistaken). I also found this thread, but I seem further along than the poster there.
If I was to prepare myself a curriculum, how quickly could I reach my goals as I stated them above? Which books should be in that curriculum? I am in Montréal, are there essential university classes in town I should take? Do you think it would be possible to attend as a vagabond student? Do you think I could find myself a mentor on Craig's list, perhaps in exchange for some programming lessons? How else could I avoid getting an engineering degree?
Thanks everyone.
I am pretty good with electronics. I have the basic circuits mastered, and I know much more about the underlying physics than I will ever use. I can even work with 120 volts without electrocuting myself (so far, knock on wood). My last project consisted in salvaging components from dead computer speakers to reassemble them into a diminutive amplifier and a power supply. It's simple really. It's a matter of following the reference circuits for the voltage regulator and amplifier chip.
I wonder where to go next, how to get better. I would like to become good with electronics, perhaps good enough to assemble an MP3 player and understand what I'm doing, good enough to build my own ergonomic keyboard, good enough to control robotic step motors from a computer program. I want to be able to build large amplifiers circuits, tweak them and debug them.
The normal path is to get an engineering degree in electronics. But I already have a master in computer science, and I am 29. I don't relish spending another four years in school. I have found the lectures on electronics in MIT open content, and I have found the online version of "fundamentals of electrical engineering and electronics". They seem to focus on high-level understanding, which I already have, and not on the practicalities of building cool gadgets (perhaps that assessment is mistaken). I also found this thread, but I seem further along than the poster there.
If I was to prepare myself a curriculum, how quickly could I reach my goals as I stated them above? Which books should be in that curriculum? I am in Montréal, are there essential university classes in town I should take? Do you think it would be possible to attend as a vagabond student? Do you think I could find myself a mentor on Craig's list, perhaps in exchange for some programming lessons? How else could I avoid getting an engineering degree?
Thanks everyone.
Best answer: Engineering is a big word, like math. Some parts of it are common to the subspecialties, and if you want to be an 'engineer', formally studying them would be worthwhile (either at a university or independantly).
If you just want to be a supertech, you can do what many techies do... pursue some aspect of the field that you are interested in and gradually reduce your ignorance as the need arises to overcome it. The choices are endless, but I have a lot of peers who have CS degrees, as do I. Where I live, getting a formal designation as a professional engineer (P.E.) does NOT require an engineering degree... it requires experience and passing the PE exam. You have to be recommended by other PE's, but the degree is not the issue. (Same is true for law and architecture here, BTW.)
The field is SO HUGE, that you can easily specialize in stuff that other folks don't pursue as much and you can get decent jobs in industry based on your skill set.
The best programmer at our shop studied music (he's a saxaphonist). The worst engineer I ever worked with was from Georgia Tech. The least socially competent was a graduate of MIT and RPI. Most of the EE's I work with can't understand anything I say about radio, DSP, or various biomedical areas that are my expertise. As a matter of fact, I have been doing this for 30 years and half my time is spent correcting their errors that I consider fairly basic stuff.
Of course, there's nothing wrong with getting that EE either, so have fun either way.
posted by FauxScot at 10:35 AM on November 26, 2006
If you just want to be a supertech, you can do what many techies do... pursue some aspect of the field that you are interested in and gradually reduce your ignorance as the need arises to overcome it. The choices are endless, but I have a lot of peers who have CS degrees, as do I. Where I live, getting a formal designation as a professional engineer (P.E.) does NOT require an engineering degree... it requires experience and passing the PE exam. You have to be recommended by other PE's, but the degree is not the issue. (Same is true for law and architecture here, BTW.)
The field is SO HUGE, that you can easily specialize in stuff that other folks don't pursue as much and you can get decent jobs in industry based on your skill set.
The best programmer at our shop studied music (he's a saxaphonist). The worst engineer I ever worked with was from Georgia Tech. The least socially competent was a graduate of MIT and RPI. Most of the EE's I work with can't understand anything I say about radio, DSP, or various biomedical areas that are my expertise. As a matter of fact, I have been doing this for 30 years and half my time is spent correcting their errors that I consider fairly basic stuff.
Of course, there's nothing wrong with getting that EE either, so have fun either way.
posted by FauxScot at 10:35 AM on November 26, 2006
Best answer: The people I know who have high level circuit building skills are mainly self taught. FauxScot's "supertech" description is right on. As you mentioned, finding a person to teach you something is helpful too. Finding someone you'd describe as a "mentor" will probably be difficult. You're more likely to find a group of other people with the same interests with whom you can exchange information.
It sounds like you have a good start. Why not jump into the bigger projects? Order the MP3 player parts and start building. See who you meet on the way and try to learn from them.
posted by nonmyopicdave at 12:02 PM on November 26, 2006
It sounds like you have a good start. Why not jump into the bigger projects? Order the MP3 player parts and start building. See who you meet on the way and try to learn from them.
posted by nonmyopicdave at 12:02 PM on November 26, 2006
IMHO, the best book for learning how to design and construct circuitry is "The Art of Electronics" by Horowitz and Hill. The 2nd edition was published in 1989. Still very useful - that's how good it is. I know lots of EEs who move to software, but not many that move the other direction. On the other hand, I once had an Engr Physics guy tell me that EE was easy, "all you guys do is look at app notes." Good luck to you.
posted by takenRoad at 8:50 PM on November 26, 2006
posted by takenRoad at 8:50 PM on November 26, 2006
takenRoad beat me to it. horowitz & hill is the definitive electronics book that i will never, ever get rid of.
as far as being good at it in practice, well, that is something of an art and requires a certain talent as well as experience to know what works, what doesn't work, what component to use here and there and so on. following instructions online is a good way to start. there is no substitute for having done it a lot.
you might also consider sitting in on some of the first-year EE courses at mcgill, which is a very good school in the sciences and engineering. i've never met a teacher who didn't let interested people sit in on their lectures, and the first- and second-year courses are probably large enough that you could just show up and not be noticed. (lab courses are another thing entirely, and you'd probably have to be a student to take them. mcgill also might let you take classes as a part-time continuing-studies student; i don't know what their policy is.)
finally a good option might be a vocational or tech school, where the focus will be less on the differential equations of circuits and more on practical knowledge. (see RustyBrooks' answer as to why.) you might think "ew, tech school!" but seriously, that sounds like what you're looking for.
anyway, sounds like you're embarking on a fun thing, so good luck!
posted by sergeant sandwich at 10:22 PM on November 26, 2006
as far as being good at it in practice, well, that is something of an art and requires a certain talent as well as experience to know what works, what doesn't work, what component to use here and there and so on. following instructions online is a good way to start. there is no substitute for having done it a lot.
you might also consider sitting in on some of the first-year EE courses at mcgill, which is a very good school in the sciences and engineering. i've never met a teacher who didn't let interested people sit in on their lectures, and the first- and second-year courses are probably large enough that you could just show up and not be noticed. (lab courses are another thing entirely, and you'd probably have to be a student to take them. mcgill also might let you take classes as a part-time continuing-studies student; i don't know what their policy is.)
finally a good option might be a vocational or tech school, where the focus will be less on the differential equations of circuits and more on practical knowledge. (see RustyBrooks' answer as to why.) you might think "ew, tech school!" but seriously, that sounds like what you're looking for.
anyway, sounds like you're embarking on a fun thing, so good luck!
posted by sergeant sandwich at 10:22 PM on November 26, 2006
The normal path is to get an engineering degree in electronics.
There are fundamentals that you will miss, but what you really need is experience. One great thing about working for a company is that they will pay for all the stuff you blow up while you are learning. The other great thing about working for a company is the experience of the guys around you that rubs off.. Getting experience on your own, as a hobbyist, seems like a pretty hard grind (especially hard on your bank account), but really, just start building and see what happens.
If you want to learn some fundamentals, and you don't feel like going over ohms law again, try hitting a book like Sedra & Smith (a pretty good book, but I'm really only mentioning it because that is what I learned from). From there, the fundamentals start to get pretty specialized. Do you want to learn analogue, digital, switch mode power supplies, microwave circuits?
The other thing you probably get, but might not.. Electrical engineering is as much about signal analysis as it is about circuits, and waves. Signal analysis meaning topics like Control Systems, Digital Signal Processing, and Information Theory.
posted by Chuckles at 10:25 PM on November 26, 2006
There are fundamentals that you will miss, but what you really need is experience. One great thing about working for a company is that they will pay for all the stuff you blow up while you are learning. The other great thing about working for a company is the experience of the guys around you that rubs off.. Getting experience on your own, as a hobbyist, seems like a pretty hard grind (especially hard on your bank account), but really, just start building and see what happens.
If you want to learn some fundamentals, and you don't feel like going over ohms law again, try hitting a book like Sedra & Smith (a pretty good book, but I'm really only mentioning it because that is what I learned from). From there, the fundamentals start to get pretty specialized. Do you want to learn analogue, digital, switch mode power supplies, microwave circuits?
The other thing you probably get, but might not.. Electrical engineering is as much about signal analysis as it is about circuits, and waves. Signal analysis meaning topics like Control Systems, Digital Signal Processing, and Information Theory.
posted by Chuckles at 10:25 PM on November 26, 2006
Hmm.. And Electronics is as much about physical implementation as it is about circuit theory. Board layout, part tolerances and temperature sensitivity, thermal management - stuff you never learn in classes. Those considerations are made much more critical when manufacturing cost is a factor (you can't just use np0 caps - interesting looking capacitor tutorial - and 1% resistors everywhere, when you are building 10,000 units), but they are important for hobby work and one off designs too.
posted by Chuckles at 10:41 PM on November 26, 2006
posted by Chuckles at 10:41 PM on November 26, 2006
Horowitz and Hill is often recommended as a beginner book, but if you haven't solved circuit diagrams with transistors, you might find it difficult. In the first 15 pages they cover what many intro college-level books cover on basic circuit theory.
posted by miniape at 10:45 PM on November 26, 2006
posted by miniape at 10:45 PM on November 26, 2006
There is also this previous question: Learning about electricity, the bright and colorful but not completely dumbed-down way?
posted by Chuckles at 10:47 PM on November 26, 2006
posted by Chuckles at 10:47 PM on November 26, 2006
I came in to plug Horowitz and Hill only to find it's been mentioned three times already. I'll agree with miniape to an extent though. I found the thing almost useless in university, and wondered why the hell I'd wasted $110 on it. It's about the only text book I own that still gets dragged out from time to time though. Once you have a certain amount of knowledge its great and its the first thing I reach for when I find myself designing a circuit.
posted by markr at 4:28 AM on November 27, 2006
posted by markr at 4:28 AM on November 27, 2006
http://www.ibiblio.org/obp/electricCircuits/
Entire set of pdf/html textbooks. Free.
posted by IronLizard at 10:23 AM on November 27, 2006
Entire set of pdf/html textbooks. Free.
posted by IronLizard at 10:23 AM on November 27, 2006
« Older Help me find a decent (but relatively inexpensive)... | Can you 'snipe' a toll-free number? Newer »
This thread is closed to new comments.
I've seen some programs on the web for informal 6-8 week classes (once a week) that look at, say, the design and construction of tube amps, and you actually build a small class A amp at the end. I haven't seen one in person but there may be things like that in your area. Leisure-learning or something, I don't know.
Anyway, good luck and I hope you find some way to expand.
posted by RustyBrooks at 9:41 AM on November 26, 2006