Hello, I was wondering if anyone had any good resources to learning about circuits using intuition instead of formal math. Actually, if it had the math too, that would be fine, but my current understanding of circuits is absolutely horrible. I don't know how to create anything, only copy what I have memorized. Even then, I don't know what good these things are that I have memorized. What is a current mirror good for, what good is it knowing where the 3dB point is, etc.?

I have been studying out of microelectronics by sedra and smith.

I would like to stay in the field of electronics because I still think it can be interesting, but I don't understand anything. Sure, I can do all the math with some effort, but I don't understand the big picture.

What can I do to improve my understanding??

I think I am looking for a qualitative description instead of a quantitative of circuits.

Seasoned circuit designers: How did you get where you are at? Can you offer me any advice or share stories of how you started?


Thanks so much.

 

Surely this is true of any pursuit or human endeavour, whether in science, engineering, arts and so on.

You can provide a person with paints, canvas and brushes but it will take years of training and trial and error for them to become an accomplished artist. They will need good teachers, mentors and critics to help them to this level of expertise. They will experiment with ideas and styles that other recognised artists have developed. They will often refine or modify what they have learned from others and develop their own personal styles. They will impose their own innate creativity as their confidence and skill grows.

Probably most of what electronics "experts" use in their daily work will be based on well established principles and practices. You may have been taught the design principles of a simple transistor amplifier. And you can solve [analyse] example amplifier problems. But it's not until you have to synthesize an amplifier from a given set of previously unseen specifications that the hard work begins - often through trial and error and guidance from others. As long as you have the basic design principles embedded in your thinking you will make progress.

It's unfortunate that some of what we are formally taught is passed to us in the absence of clearly stated application examples. A reasonable question from a good teacher's students should always be "What is a typical application for this?"

 

can you recommend resources for me? (preferably textbooks or inexpensive projects - Budget considerations)

I have no problem teaching myself things by resource hunting and critically thinking on my own. For instance, I taught myself everything I know about Chinese, Spanish, Piano, guitar, and singing, but I am having a big problem with electronics. The math is not the problem. I can follow along with the math. If I could sum all my complaints into a single phrase, my complaint would be that I don't understand what good any of this stuff does / how it applies.

 

I had the same problem with the same book.
I thoroughly hated it when I first cracked it open.

2 years and more advanced electronics courses later, I went back and did a lot of the Sedra / Smith problems.
It's a really good book with a lot of information but only if you're used to it.

Have you done the chapter 1 problems at all? I know my course skipped it.
It's very useful because it's a higher level look at just generic amplifier models, pretty much just input / output resistance.

What it gives you though, is an idea of what you're aiming for.
Having a goal is always crucial for me for understanding.

So for example, a realistic problem would be that you have an audio source with some output resistance and a certain voltage level.
You then have a speaker that you want to drive a certain voltage level to get the right amount of volume.
That's what you're starting with.

How do you connect it?
Your source most likely cannot be connected directly to the speaker so you need some amplifier in there.
That amplifier will need a specific gain and specific input and output resistances to work properly.

If you try using a single transistor amplifier you will probably find that you can't get everything you want, either the gain is too low or the resistances eliminate your gain.
So then you can add a second stage, or maybe pick a different kind of amplifier.

Once you have a basic idea you can start tweaking the circuit. Bias current(like from a current mirror) will change your gain, and by tweaking your resistors you will probably change the gain as well. It's all about trade offs between these things.

Since this is audio you will only need about 20 kHz of bandwidth, so you need to have your 3 dB frequency around there or beyond it. Again, you will have trade offs between that and gain and your resistances.

None of it makes sense until you see the goal...

Seriously though, try the problems in chapter 1, then try problems from the other chapters. They're really helpful, especially the design problems. Just do the ones with answers in the back and especially the worked examples inside the chapters. Try doing them without looking at it.

I'm not exactly seasoned by the way, I just graduated electrical engineering last year. Still haven't built much...

 

When I was young - many years ago - my classmates & I used to read electronics magazines and occasionally built the circuits that were of interest. Some were interested in music so they built electronic circuits for guitar special effects, etc. Others were interested in radio communication so they built RF circuits like VFO's or played with antenna designs.

I would reckon that some of them went on to design & build their own circuits once they understood what the various circuit "building blocks" did.

A lot of circuit design is about knowing what are the basic building blocks of a system & what are well established design techniques. I would suggest one of an electronics engineer's / technician's strengths is the ability to apply prior knowledge with those modifications that are necessary to meet a particular design specification in a reliable and cost effective manner.

Very few engineers start out as design experts. No company or organisation would entrust a recent graduate with the sole design & management of a new electronic system. Over time they will be assigned tasks commensurate with their demonstrated ability and successes.

Are you currently working in the design field or are you still studying?

 

can you recommend resources for me? (preferably textbooks or inexpensive projects - Budget considerations)

I have no problem teaching myself things by resource hunting and critically thinking on my own. For instance, I taught myself everything I know about Chinese, Spanish, Piano, guitar, and singing, but I am having a big problem with electronics. The math is not the problem. I can follow along with the math. If I could sum all my complaints into a single phrase, my complaint would be that I don't understand what good any of this stuff does / how it applies.

There aren't any easy ways to do what you want to do with that methodology. It is a complete waste of your time. Not everybody is cut out for this stuff.

 

There aren't any easy ways to do what you want to do with that methodology. It is a complete waste of your time. Not everybody is cut out for this stuff.

Can you clarify what you mean? I thought I had a pretty good methodology? Can you give me some further tips? What should I do instead if this is a waste of my time?

I like how you disagree, but I don't like how you don't offer any better alternatives.

 

I would be very surprised to find some material that said it taught electronic design by intuition. It would be the other way around - once you have a body of knowledge and experience, then intuition may form a partial guide to your designs. You will be able to eliminate any number of unrewarding approaches when beginning a new design.

That same body of experience is very helpful in evaluating a design from some source other than yourself. With enough background, troubleshooting a problem goes as much by intuition as by some arbitrary checklist function.

The best way to get good at something is to put the knowledge from the book to use.

 

Intuition:
b. Knowledge gained by the use of this faculty; a perceptive insight.

Intuition is something that happens after years of twiddling with a variety of circuits.

Once the "Big Picture" of different circuits, their functions, and known shortcuts for some of those functions is known, looking at simpler circuits, like many posted here, are seen by many of the seniors as "Intuitive", and by the people new to electronics as black magic.

Without the math, behavior of devices, and electronic laws not only learned, but embedded into your brain, I cannot see "Intuitive Circuit Analasys" possible.

 

It is very hard to learn electronics from scratch just to learn it.

I think most have a goal set.
"I want to build a car that follows a line!"

Then you break it into a flow-chart.

Now you know what needs do be done, and you can research the components that do those things.

For some people, the water analogy is the only way they "get it".

 

For some people, the water analogy is the only way they "get it".

And then they ask about AC!

I've been in electronics for 30+ years, and in that time I've worked with lots of different people, in various capacities. What I have learned is that, beyond even a fairly rudimentary level, the speed of the progress tends to be determined by a natural instinct.

Bit like music or painting I would think.

I have met graduate engineers I wouldn't trust to mend my kettle, and at the other end of the scale, a 'natural' who only had to have someting described once, and would then nearly always grasp the concept, and was an excellent test and repair tech.

Interestingly, his maths wasn't good, so he never climbed very far up the food chain, without the paper qualifications to go with his instinct.

I started by learning to repair electronic organs. In the 1970's these were huge expensive lumps which had to be fault found, to component level, on site. So the training and expertise of the service techs was much higher than it is today. (not the salaries though! )

You learn to sink or swim fast in that kind of world --you either 'get it', or you don't!


At the end of the day though, nothing beats the value of experience. I can still remember the first time I designed my own circuit from a blank sheet of paper, and got it to work. There were lots of times before that where it didn't -- so you want to know why, and try again, and again, and again.........
In those days, there were no 'sim' packages -you just built up a breadboard, and tried different things until the smoke stopped coming out!

It's interesting to see how trends ara changing though. Almost no hardware 'hobby' magazines (although I believe Elektor is still going??). Nearly everyone seems to start their project with the words 'so I know I need a PIC'. It's a different world!

I still like the idea of starting with a simple hardware project, getting hold of some breadboard, a cheap secondhand 'scope, a current limited power supply, and an a few components, and just physically build something.

Now that doesn't work......why?....lets try this........ah! that's a bit better..........and so on!

 

Venficus5, If you me your ear I will make you feel better about learning
electronics.I will be waiting for your post,I will answer.

 

Can you clarify what you mean? I thought I had a pretty good methodology? Can you give me some further tips? What should I do instead if this is a waste of my time?

I like how you disagree, but I don't like how you don't offer any better alternatives.

I don't need to offer an alternative, because it is the one you are already aware of and have rejected. You are wasting your time with your new approach, in my opinion, because it has an exceedingly low probability of success. I also don't care what you like and don't like. You're the one that put the idea out there and said in effect "look at my baby, isn't she beautiful". Then you take umbrage when someone says "Sit down, your baby is ugly". In a standards committee I once served on, we called this phenomena "The Ugly Baby Bus". Every company on the committee had its parochial interests and agreement on even good ideas was a hit or miss proposition.

If you're not prepared to deal with contrary ideas and opinions; it is hard to see the value of advancing them.

 

Even then, I don't know what good these things are that I have memorized. What is a current mirror good for, what good is it knowing where the 3dB point is, etc.?

If you don't have a problem to solve, there is no point. Circuits are building blocks. If you want a LED to shine, you look around and you only have a 120vAC power source, you now have a need. You need a circuit to rectifiy the AC to DC, and you need to limit the current to the LED with a resistor.

You can use the memorized circuits for the building blocks.

The reason is problem solving. If you have no problem to solve, there is no reason.

 

Venficus5, I have some southern comfort for you don't give up yet.looking
forward to hearing from you ,out the door for another great adventure in life.
Don't go far "I shall return to make good on my word". Open your mind and I
shall fill it with good things about electronics. I am the the only -----------

 

Start with trying to understand the most basic components like resistors, capacitors, diodes etc.

Until you grasp what each does and can be used for, you will not properly understand anything futher, it will just be jargon.

It may seem tedious, but it's the only way to proceed and the only way you will be able to follow circuit diagrams.

 

ok maybe I chose the wrong title to this thread. Of course it can't be done solely by intuition. I would like to have conceptual understanding as well as the algebraic understanding I already have. I believe this can be obtained through the use of abstracting complex circuits down into recognizable/pre-memorized blocks, but being a beginner, I don't know that for sure. The problem is that I have yet to see this put into use. Am I just not at a high enough level to start doing this yet? Sort of like a student in algebra? Algebra is completely useless on its own until it is used to solve problems.

see attachment picture
For example, in physics, say you have a problem where you want to know with what force to pull on a rope north to get a kite to point in a certain direction given a certain wind speed east.

Without getting bogged down in the details, you know that the harder you pull the kite north, the less the angle will be. Assume the kite can stay afloat without any pulling. If the kite is pulled down, then the mirror situation will occur. If the Kite is not pulled on at all, then the angle will be 90 degrees.

Is there any way to do this with a circuit? Well... I know there is, because why would we worry about building blocks and layers of abstraction if it weren't possible to get a general idea. So my question is, what resources use this approach? To me, the sedra and smith book just throws a bunch of random crap together and never builds anything useful with it. There is no motivation if the stuff can't be applied. Obviously it can be applied if there is a degree titled electrical engineering, but I am just lost for finding the applications.

Thanks!!

 

If wishes were horses, then dreamers would ride.

Just because you want something to be so, doesn't mean that it is or can be. Knowing what something is good for does not come at the beginning. It comes by listening to other people and even working on their ideas. It was probably 6 or 7 years into my career before I had my first inspirational idea. I spent most of those formative years after getting out of school (5 years) absorbing everything I could in my environment. I even spent 6 months just thinking and learning about CRC algorithms to see if it was possible to compute one on a limited, non-microprocessor ALU, in parallel instead of 1-bit at a time. It is possible, as is well known today

This was decades before the internet made access to information a great deal easier than it is today. I never thought to ask what good this stuff was. My company needed it and that was all there was to it.

 

These have replies have a theme of "you are lazy" or "you are stupid". The latter is true, but I am not lazy. I just have no motivation to learn any of this stuff because I see no application. I have no problem learning when motivated either internally (Fun ~ Spanish, Chinese, Python, Guitar, Piano) or externally (Salary ~ Matlab, C++, writing, math, various aspects of the internal motivations above, etc.).

To me, the applications of the EE I am learning are not obvious. It is somewhat interesting, but if it doesn't pay, there are plenty of things much more interesting. I guess I don't see how knowing this stuff would benefit a company and in turn benefit me. That is the reason I am considering software engineering. To me, the applications of software immediately jump out. I use software every day, I browse the internet, I use applications, etc.

It seems like this analog stuff is going out the window. Please give me examples of where it's not!! If there are still applicable uses, I would love to stay in the field. For instance take filters. Can't a digital filter through software do everything an analog filter can and more?? Or Signal processing, or an audio synthesizer, etc.

Maybe it is because I haven't taken enough classes yet. To give you an idea of where I am at, I have taken three circuits classes (Ohms Law, KVL, Phasors, Power, Laplace Transform, Fourier Transform) and two electronics classes (Basic knowledge of BJTs, MOSFETs, current mirrors, differential amplifiers, and a little on digital IC design)

Thanks so much. As much as I am trying not to, I realize from your replies that I am coming off as a complete moron and know-nothing kid, but that's exactly what I am; I am seeking guidance of you people that know so much more than me. Sincere thanks!!

 

Analog cannot go away because the world is analog.

Use a sensor and you need an analog circuit to properly pass it to an ADC.
Use high resolution ADC's and you need to worry about everything because millivolts matter, sometimes even microvolts.

You can't just sample everything, you'll need anti-aliasing filters which have to be analog.

Go to high performance digital circuits and you get into analog problems - non-ideal capacitor behaviour, transmission line effects, crosstalk... with crosstalk you even need experience with electromagnetism because it's about geometry and fields.

Even putting down a power supply for your digital circuit you need to worry about a lot of analog things.