4
\$\begingroup\$

I am a sophomore electrical engineering student majoring in electronics and communication.

I have always been an electronics enthusiast since I was 9 years old (19,now), but there was always one fundamental electronics building block that I was missing Transistors, I recently had dived the deepest inside the physics of it, I seem to understand it better but not practically at all.

I can't use it in a single simple circuit either as a switch or amplifier, 5 years struggling to understand this topic and no progress.

My question is how did you guys learned it and started using it in your projects and how do you suggest that should approach this topic to get a really good feel for how to use it in my projects?

Thanks for taking the time to read.

Edit: thanks for the great answers everyone it was really helpful to know the different ways that you learned about this topic, but i think the most common approach that I actually didn't give much time is to just make anything with it.

Edit 2 : this is not an opinion based question as some may have said, and it was answered using facts. The clear question is how a person learned something and how should others learn it , like any other topic NOT Opinions based.

\$\endgroup\$
16
  • 1
    \$\begingroup\$ Start with a common collector (emitter follower) circuit. Can you draw one and calculate bias resistor values? \$\endgroup\$
    – Andy aka
    Oct 8, 2021 at 8:13
  • \$\begingroup\$ Find a transistor project that you want to make for yourself, not because it is a circuit but because the result will be something you will want to use as a tool. Then take the time needed to apprehend that circuit well enough that you can change it and get improved results. In my day, this might have been most anything because most circuits then were in the process of replacing tubes with BJTs. But today you may have to look. But an example, which unfortunately requires also some inductor skills, might be a metal detector based on BJTs. Find something you want and care about. \$\endgroup\$
    – jonk
    Oct 8, 2021 at 8:13
  • 1
    \$\begingroup\$ I do actually think I should really get this book it isn't the first time I have seen recommendations for it \$\endgroup\$
    – AhmedH2O
    Oct 8, 2021 at 8:25
  • 1
    \$\begingroup\$ > but it didn't work and I couldn't troubleshoot it." BUILT HOW? Plugged into white prototyping block? Beginner troubleshooting: assume you made a mistake, so LOOK for errors, no knowledge or DVM needed. Triple check: was transistor plugged in backwards, or transistor pins wrong, or soldering heat killed transistor, or brief reversed voltage killed transistor, or ESD static killed transistor, or wrong resistor values. Transistors are easy to destroy, so never buy just one, be ready to swap in a new unused component, in case it was killed by accident, and now forever prevents function. \$\endgroup\$
    – wbeaty
    Oct 8, 2021 at 10:15
  • 1
    \$\begingroup\$ > also I tried EEVblog's current multiplier" those sound like very advanced circuits, where a beginner will always fail. To learn electronics, first build many many simple projects (one-transistor projects,) and NEVER give up on any. Make many mistakes, destroy lots of transistors! Find your errors, since you must learn which mistakes are common. After years, after success with many simple projects, then the simple projects will seem easy, and will always work. Only then should you try a complicated one. \$\endgroup\$
    – wbeaty
    Oct 8, 2021 at 11:11

5 Answers 5

10
\$\begingroup\$

Before I had a formal education in electronics, my understanding of transistors was really sketchy. When I started courses it was immediately clear why I couldn't grok them properly - I didn't know about current and voltage.

I mean, I knew what they were, and I knew Ohm's law, but I didn't have a good enough understanding of how currents and voltage add and subtract, what happens at branches, or how they were all related in a bigger picture sense. I thought current got slowed down at each resistor. I thought current couldn't flow through a capacitor. I thought a bunch of things that were just plain wrong, because I read them in books for beginners or magazines, or got taught by people who didn't really understand either.

Everything I had read was really explanations for 10-year-olds, and completely innapropriate for anything more complex than an LED, resistor and transistor setup.

My problem, it turned out, was never with transistors, it was with Kirchhoff's Voltage Law, Kirchhoff's current law, and my fundamental understanding of what a voltage source is, and what a current source is. Terms like "sinking" and "sourcing" current were missing from my vocabulary. Lots of little things were missing.

Transistors are hard, it's true, but they are impossible to understand without a really solid foundation to place them on in your mind. In my case I had to unlearn a lot of crap, and luckily I had university to re-educate me. These days there's the Internet, but I imagine it's hard for a beginner to identify the crap from the gold in all those Youtube videos. Some of them make me cringe.

Anyway, there are thousands upon thousands of "exercises" out there, and you should do as many as possible. Every day you'll find something that either fills a gap in your understanding, or highlights some misconception you aready have. Believe me, even today I still have "aha" moments about stuff I thought I had figured out 20 years ago.

Let me highlight some issues that I (and my students) have had to address. They might not apply to you, but they might help.

Vernacular

All students of mine use terms that highlight problems with their understanding. Perhaps you can identify your own problems by trying to find out why certain terms are used.

When a student says to me "the voltage through this resistor" I immediately know they haven't understood what voltage is. Voltage doesn't flow through things, voltage exists at places, or more precisely voltage difference appears between places. I do not use the word "appears" for no reason, it implies something.

When they say "the current gets to this resistor and ...", I can be almost certain they've misunderstood the nature of current. In a long chain of components, the current is same everywhere along that chain, independent of the order of the components, or their number, or time. Voltage and current are established instantaneously everywhere in a circuit, and are a function of everything in the circuit, all at once. This becomes clear when you start solving the simultaneous equations from Kirchhoff's and Ohm's laws. They are called simultaneous for a reason.

"At" and "across", are terms used to refer to voltage only. As I already stated, only voltage differences exist, between two points, and absolute voltages do not exist. Therefore, when someone says "the voltage at point X is 15V", what they really mean is "the voltage between point X and some point I've arbitrarily declared to be 0V (ground), is 15V."

Laws

Learn your basics really really well, and the transistors will fit in just fine. As I said, do exercises, and in particular look for these terms:

  1. Kirchhoff's Current Law
  2. Kirchhoff's Voltage Law
  3. Ohm's law and the Power Law
  4. Thevenin (and perhaps Norton) Equivalents

Potential dividers

95% of circuits rely on potential dividers (a.k.a. resistor dividers, or voltage dividers). Learn to spot them, because they are everywhere. Understand what happens to all the current as you increase this resistance here, that current there, tweak this voltage here and so on. Understand them in terms of the current through the components, the voltage across each component.

Develop an intuition for their behaviour, until you no longer have to think about it. Whenever you see two or more resistors in series, think "potential divider" and can instantly see how the total voltage is shared across each, to the point where you can estimate the voltages just by glancing at the resistance values. It's not as hard as you think.

Even a transistor can be a member of a potential divider, and it has to obey the same rules as every other member. If it becomes "more resistive", the voltage across it and current through it will do the same as a resistor would in its place. If you understand what resistors in a potential divider do, and how the voltages and current change when an indiviual resistance changes, then replacing one with a transistor is not such a leap of understanding.

Here are a few examples, of potential dividers, some using only resistors, but some using a mix of components. Below them I'll write some comments about my instantaneous thoughts when I see them:

schematic

simulate this circuit – Schematic created using CircuitLab

  1. The resistances are identical, passing the same current, dropping the same voltage. Therefore the voltage at X is half way between the top and bottom voltages, so \$V_X = 6V\$.

  2. 100Ω is very small compared to the two 10kΩ resistors, and so the voltage it develops will be very close to zero. The voltage at Y must be near zero. The remaining majority of 12V must be divided equally between the other two resistors, at slightly under 6V each. Therefore the potential at X must be slightly over 6V.

  3. The two resistors have the same value, sharing the same voltage across them, as for (1) above. X must be half way between the top and bottom voltages, +24V and -24V respectively, which is zero volts. \$V_X = 0V\$.

  4. The LED chooses its voltage, which will be about 2V for a green model. \$V_X \approx 2V \$, irrespective of the value of R8. The remaining 3V must be across R8. By Ohm's law I estimate current through R8 and D1 to be about 10mA (3V ÷ 330Ω)

  5. The transistor Q1 is off. That means the effective resistance between its collector and emitter is very very high compared to R9, so most of the 5V source is dropped across Q1. The voltage dropped across R9 is practically zero, and the voltage at X must be 5V.

  6. The top transistor Q2 is off, so its drain-to-source resistance is very high. The bottom transistor Q3 is on (saturated), so its collector-to-emitter resistance is very low. Most of the 12V source is therefore dropped across the highest effective resistance, Q2, with almost 0V remaining across Q3. \$V_X = 0V\$

I made every one of my students do exercises like this, to describe qualitatively, and without algebra, the behaviour of divider configurations like this. It's important to be able to take an educated guess at the distribution of voltages, without resorting to arithmetic and algebra every time. You absolutely need to develop an intuitive feel for this behaviour.

Redraw the circuit

Often it's just not possible to understand a circuit because it's been drawn in a way that doesn't suit your mental model of direction of current flow, or potential "steps". In my mind, and I think in the minds of most engineers, current flows from high potential to low, in the same way that water flows downwards from high gravitational potential to low. It makes sense to draw circuits with high potentials at the top, and low potentials at the bottom. That way potentials drops as you move down the circuit, and current flows from top to bottom. Do this just to facilitate the mental gymnastics that go on when trying to analyse the circuit.

Do the same with signal directions. Left to right, just like writing. Inputs arrive from the left, outputs leave to the right.

Obviously this is not always possible. There may be feedback signals going to the left, and sometimes you don't know which potential is higher, but you can always simplify your mental processes by trying to stick with these conventions. It will also help when communicating with other engineers.

Here are two representations of the exact same circuit, the first is a mess, the second is easier to analyse:

schematic

simulate this circuit

I want to draw your attention to the resistive potential divider outlined in the orange box. It's positioned between the opamp output and ground, and is dividing the opamp's output potential by about 3. Learn to spot these "modular" elements.

Here's an example of a circuit which makes my eyes hurt, and which illustrates the importance of this:

schematic

simulate this circuit

In the left version I have to turn my mind upside down to see what's going on. When you ask me "what's the voltage at X?", I can't immediately tell you.

If you show me the circuit on the right, and ask the same question, I immediately know that at Y the potential is 6V,and at X it must be 1.7V less than that, at 4.3V, due to the drop of 1.7V across a typical red LED. And yet the two circuits are identical.

Apologies if any (or all) of that was already obvious to you, but I do hope it helps to build some skills ready for the transistors to come. These ideas have helped me (and my students) a lot over the years.

\$\endgroup\$
7
  • \$\begingroup\$ Thanks for taking the time to write a well versed answer like this one I definitely had to relate to at least some of what you said. \$\endgroup\$
    – AhmedH2O
    Oct 12, 2021 at 14:02
  • \$\begingroup\$ “ I thought current couldn't flow through a capacitor” // It doesn’t flow through the dielectric (in normal conditions). \$\endgroup\$
    – alejnavab
    Oct 12, 2021 at 18:21
  • 1
    \$\begingroup\$ @AlejandroNava, you are not wrong, but in the context of KCL (and of this question) this fact is irrelevant, and is truly misleading. Current flows, and it doesn't matter how, in the same way it doesn't matter from an analytical standpoint that electrons are the charge carriers. Both of these "truths" are irrelevant and only serve to muddy the waters. If I can push one amp into one side of a capacitor, then 1A will emerge from the other, it's as simple as that. \$\endgroup\$ Oct 13, 2021 at 2:11
  • 1
    \$\begingroup\$ @AlejandroNava if we say that capacitor dielectric has no current, then we'd also study wires, find big spaces between the electrons, then declare that there's zero current in wires! (After all, the wire is mostly empty space! Heh.) In capacitors, the current in one plate must always be identical to the current in the second plate, because the gap is small, and the dielectric forces this to happen, by physics-magic called Displacement Current. (Also in modern caps the dielectric is not a vacuum, and current in high-K dielectrics is partially-free, moving electrons. See "ferroelectric." ) \$\endgroup\$
    – wbeaty
    Oct 13, 2021 at 13:05
  • \$\begingroup\$ @SimonFitch “but in the context of KCL (and of this question) this fact is irrelevant” // Correct. I was just adding a note in case someone misinterpreted your statement; I should’ve added this GIF also. \$\endgroup\$
    – alejnavab
    Oct 13, 2021 at 20:12
3
\$\begingroup\$

We (at least most of us did) got to know transistors by doing experiments, seeing what happened, and gradually internalising an understanding of the transistor from explaining the behaviour. Book learning is rarely enough. It takes time, and repetition, and thinking through the reason for sometimes puzzling behaviour.

Here's a simple labororatory for you to use.

Build this circuit, either in real life, or in a simulator, or better still, in both. Almost any NPN bipolar junction transistor will do, whether signal or power.

schematic

simulate this circuit – Schematic created using CircuitLab

And then do some experiments on it. Use a DMM to measure voltages across various parts. Short-circuit a resistor and see what happens (any single resistor at a time can be short-circuited without damage). Remove a resistor. Vary a resistor up or down.

As a start, short R4, then measure the voltage on R3 as you omit either R1 or R2. This is the basic switching operation.

Or build the circuit exactly as is. This is the follower configuration, where the voltage on R4 follows that on R2, less a VBE drop. Short R3, and it still does essentially the same thing. Leave R3 in place, measure the voltage on R3 and R4 as you play around with the values of R1 and R2. Notice anything interesting? Why?

Short both R3 and R4, and watch the transistor burn. You were using a cheap expendable device, weren't you?

\$\endgroup\$
1
  • \$\begingroup\$ Thanks mate I will certainly try and I would definitely be happy to blow out some transistors \$\endgroup\$
    – AhmedH2O
    Oct 8, 2021 at 9:34
1
\$\begingroup\$

At first, build some simple and fun projects that always work. It doesn't have to be "useful," instead it must be something YOU want to build. Your goal is to have a functioning circuit which does something cool, NOT to understand it. (Understanding happens later, over years!) First, build thirty or fifty (or several hundred) little transistor projects. Start with white prototyping-blocks, no soldering.

You can troubleshoot without DVM meter or understanding electronics. Beginner troubleshooting: assume you made a mistake, so LOOK for errors, you know they are there, so double check everything. Triple check: was transistor plugged in backwards (and now it's killed, must replace?) Or transistor pins connected wrong, or soldering heat killed transistor, or brief short-circuit or reversed-voltage killed transistor, or ESD static killed transistor, or wrong resistor values used (don't design your own circuit! Use resistor values from a project article.)

Transistors are easy to destroy, and at the start, you're going to kill lots of them. So never buy just one, be ready to swap in a new unused component, in case it was killed by accident, and now forever prevents function unless replaced.

Don't try to design anything (you may just fail and give up! DON'T ABANDON A PROJECT, stay with it, find the error, get it working.) Be a noob beginner, if that's what you are. Stay simple, starting with just one transistor or two. Make an LED blinker. Make a sound generator with a tiny loudspeaker. Or be like me, doing physics and detect invisible voltage-fields in the air with an FET, see my article: http://amasci.com/emotor/chargdet.html

Which projects? Find all the online project archives, magazine archives, or buy older books full of one-transistor projects. Here are some online sources:

Old magazines have simple projects for beginners. Try these PDF archives:

Oh, and the book "Art of Electronics" is college level. It's worth having a copy, but it's not so great for beginners. Much better is the EVIL GENIUS series, search online booksellers for: "evil genius" gadgets, also "evil genius" projects.

\$\endgroup\$
0
\$\begingroup\$

Full understand of all transistor possibilities is a long run. Is has a planty options but I would start with basic CE connection.

First: Try to understand how the currents in transistor flow. How base current affects the collector current. (Ic = Beta * Ib) Beta temperature changes let go for now.

Second: Look at transistor chart (Vce vs. Ic vs. Ib)

Third: Try to understand the load line in transistor chart for different collector resistors. Especially the slope changes. Emitter degeneration using Re let go for now - it complitates the dependencies/ but calculations using equations isnt so hard.

4th: Understand base biasing types - Rb from Vcc / Rb from collector / Beta independent bias.

5th: Try calculate/scale the output for desired range. Also figure about the impedance match, the right Rb and Rc sellection.

6th: Write euations and find what and how affects the Beta changes. How using tr. pairs can avoid this affects.

Even after 10 years of designing transistor circuits there still be something to surprise you.

Edit: One thing that really help me in beginnings is to think the transistor CE as variable resistor.

\$\endgroup\$
1
  • \$\begingroup\$ Thanks mate I like the edit you added \$\endgroup\$
    – AhmedH2O
    Oct 8, 2021 at 9:34
0
\$\begingroup\$

I want to share my experience (from the point of view of someone without any formal electronics training).

  1. I started using op-amps for most my analog circuits because they are extremely easy to understand.
  2. My contact with transistors was for two purposes that I couldn't achieve easily with op-amps: more output current than op-amps and small switches or - more generally - variable resistors.

My suggestions is start with MOSFETs since IMO these are much simpler than BJTs to understand. To boost output current of op-amps a simple source follower configuration is extremely easy to do, and for switching they have a pretty nice trigger point. It is probably useful to focus only on N-MOSFETs first and then when you are adept, look at P-MOSFETs.

Then when you are good with MOSFETs, consider the following statement: it is very often possible to interchange a MOSFET for a corresponding BJT. The advantage of the BJT are: lower cost, lower turn-on voltage, lower parameter spread, lower noise, larger transconductance, and it can be used bipolarly whereas most MOSFETs are only useful in one direction. The disadvantages of BJTs are lower current capability and a "leaky gate" (aka "base"). If the advantages outweigh disadvantages you can use BJTs (which frankly happens quite often).

So in a nutshell, get familiar with MOSFETs and then think of BJTs and MOSFETs as slightly different flavours of the same thing.

\$\endgroup\$
4
  • \$\begingroup\$ Doesn't Mosfets work on different principle using voltage instead of current? \$\endgroup\$
    – AhmedH2O
    Oct 12, 2021 at 14:00
  • 1
    \$\begingroup\$ @AhmedH2O They do work by different principles, but Mosfets work by voltage-input, and Bipolar transistors also work by voltage-input! BJT transistors aren't current controlled, that's a "Lie to Children," see Art of Electronics text for in-depth explanation. Search "Ebers-Moll." Yes, we teach kids that BJTs are "current controlled," since an oversimplified lie is far easier to teach than a complicated truth. BJT transistors are controlled by the potential-barrier in the EB junction: by an exponential voltage/current function, same as any diode. We only pretend that Ib can affect Ic. \$\endgroup\$
    – wbeaty
    Oct 13, 2021 at 13:13
  • \$\begingroup\$ @AhmedH2O yes the construction of a MOSFET and physical operation principle is rather different. However, as wbeaty nicely explained, from a user point-of-view, the N-MOSFET and NPN BJT are actually rather similar. You can think of BJTs as MOSFETs with very low capacitance and very low threshold voltage or conversely, you can think of MOSFETs as BJTs with extremely high \$\beta\$, but worse transconductance. \$\endgroup\$
    – tobalt
    Oct 13, 2021 at 13:30
  • \$\begingroup\$ ok thanks, that makes sense . \$\endgroup\$
    – AhmedH2O
    Oct 13, 2021 at 17:37

Not the answer you're looking for? Browse other questions tagged or ask your own question.