# The 'gap' between understanding of transistors and their application in real circuits

How do they even work? I'm in senior high school year and have electronics as a subject. I'm really interested in getting to understand this and take up electronics in college too. But now this seems like a distant dream with my feeble understanding about 'how transistors work' and their 'actual applications in circuits'. I've read numerous guides online and after finishing them, I feel like I've learnt most of it but when I start studying about the TTL NOT gate (IC 7404) and a few others, (like 7402, 7400) which is part of my coursework, and which is based upon working of the transistors, I don't get anything! Sometimes the emitter is used as input, sometimes, it is used as output and I feel some of the sentences in the text (about the working of ICs) contradict what I learnt in other guides. I feel there's some gap remaining between, my understanding of transistors, and their applications in circuits, which is pissing me off and can't seem to see what it is.

Can anybody please point out some articles which would fill this gap and enlighten me?

Update: I'd like to learn about their working in application circuits. About the 'depth of understanding', I know what role electrons and holes play in working of transistor.

• Also, I mentioned TTL NAND gate... Does it not necessarily have a totem-poll output stage? – kapeels Nov 29 '11 at 22:17

Buy this book The Art of Electronics by Horowitz and Hill (2nd edition).

It cost \$US20 (which is a bargain). It's in New Delhi and they have a number of them. If you cannot afford the 1050 Rupee get several friends to buy it together, This is the best book on the subject that you will find.

• The Art of Electronics (Second Edition)
(ISBN: 0521689171 )
Paul Horowitz,Winfield Hill
Bookseller: BookVistas (New Delhi, DEL, India)
Bookseller Rating:
Quantity Available: > 20

WARNING" There are a lot of these also advertised in India. They cost typically the same or more as what I recommended and are not the same. Take due care. This the associated student manual by Horowitz and Hayes. If you can afford to buy one of these AS WELL do so but get the proper textbook first. Copy of workbook here for Rs484 including postage in India.

• This is the book which is always recommended, but honestly I think it's a poor recommendation. It goes very very in-depth, which makes it a difficult read even for someone with past experience with electronics. Also, I read technical books almost exclusively, and even I found the writing to be dry. I would hold off on reading this book until you are more experienced. – BlueRaja - Danny Pflughoeft Nov 29 '11 at 19:40
• @BlueRaja-DannyPflughoeft - I'd have to disagree with you there, IMHO the lucid writing style and careful attention to detail is probably one of the main reasons this book stands out from the rest. – Oli Glaser Nov 29 '11 at 21:57

Do you want theory of how transistors work at the semiconductor level? Or just practical application stuff? If it's the former, I don't have much to recommend there... it's terribly complicated stuff, and supposedly requires at least some knowledge of Quantum Mechanics to fully understand. But in terms of simply using transistors, I found the Make:Electronics - Learn By Discovery book to have a nice introduction.

http://www.makershed.com/product_p/9780596153748.htm

Beyond that, I'll just share these points from my own experience: Think of a transistor as a switch, where the resistance between two of the "legs" (collector and emitter, OR drain and source in the case of a MOSFET) can be varied based on a signal applied to the other leg (base, OR gate in the case of a MOSFET). People say transistors "amplify" a signal, and that's misleading to some people's intuitive understanding. They amplify the signal to the base/gate in the sense that the base/gate controls the current flowing through the other two legs, but there has to be power supplied from somewhere in the first place. That is, they don't magically manufacture current (or voltage).

Soooo... if you have, for example, a 12VDC power supply, with a lead going from the power supply, to the collector of a transistor, and then a lead from the emitter to a load, and then to ground... a smaller signal (at, say, 5VDC) controls the current to the load. So, in a sense, you can say that that smaller signal has been amplified.

At other times, you don't really care about any sense of "amplification." You just want something to switch on or off, so you can implement binary logic. So if you think of "off" as a binary "0" (or "false") and "on" as a binary "1" (or "true") you can work out how transistors can implement any arbitrary bit of digital logic.

When you start talking about ICs like the 7400, 7402, 7404, etc., think of them as just prepackaged bundles of transistors that implement some particular bit of logic, that you can use as a modular building block. You could wire up a NAND gate, for example, by hand with a couple of transistors. But using a 7400 series NAND gate is simpler because it's already built for that purpose. Progressively more sophisticated ICs feature more and more transistors to implement more complex functions.

• If you do really want to understand the semiconductors, the subject you're looking for is solid state physics, which indeed builds off of quantum mechanics. If you're interested in a deeper understanding, I suspect recommending my college textbooks would be unhelpful, but you could perhaps ask for recommendations on the physics stackexchange, clarifying your current level and what it is you hope to learn. – Cascabel Nov 29 '11 at 20:16
• "Think of a transistor as a switch, where the resistance between two of the "legs" can be varied based on a signal applied to the other leg" You mean a variable resistor, not a switch. When only the ends of the variable resistor are used, it behaves like a switch. Agreed 100%, though; it's a controllable valve, not an amplifier. – endolith Jun 24 '14 at 13:07

Transisters, when used in digital circuits, work as electronic switches. And with switches, we can create logic gates.

Take a look at the following diagram:

If we call +VDD "ON" and ground/0 "OFF", then when the switch is closed, the output is OFF; and when the switch is open, the output is ON. If (like with transistors, as we'll see in a minute) we call a closed switch ON, then this circuit is an inverter: when the input is ON, the output is OFF, and vice-versa.

If we add a second input in series, we now have a NAND gate:

Since it is well known that all logical circuits can be built using only NAND gates, we now have the ability to build any logic circuit.

Here is what the equivalent circuits would look like using transistors:

The fact that computers essentially require nothing more than simple switches explains how computers existed before transistors - they can be built using vacuum tubes, or relays, or even normal physical switches.

In fact, you can even build a working computer out of redstone or dwarves ;)

• -1 - Transistors, when used in digital circuits, work as electronic switches No, they really don't. In a computer or digital device, a subset of the analog range of a transistor is treated as a binary signal. However, it is still absolutely analog, with turn-on and off still involving a voltage slewing between the two "binary" states. – Connor Wolf Jan 26 '13 at 10:15
• You have to be careful thinking of discrete transistors, or even logic ICs in purely digital terms. It's all analog under the covers, and ignoring this fact will come back to bite you. – Connor Wolf Jan 26 '13 at 10:16
• @ConnorWolf For someone struggling to understand transistors, yes, they are switches, and going into any more detail than that is only going to confuse them. Real switches don't instantaneously change from off to on either, but those kinds of details don't matter 99% of the time. – endolith Jun 24 '14 at 13:14
• @endolith - That's ok, but then you should say "can be modeled as electronic switches", or something similar. Simplified explanations are fine, but they should be explicit that they're just simplified explanations. – Connor Wolf Jun 24 '14 at 13:16

KPL, I completely understand your frustration. It sounds like the problem you are running into is the question of what is occuring inside the material of the transistor itself. Remember that a single transistor is simply a switch that turns on and off in response to the presence of voltage on it's "third" input. A voltage there causes the switch to close. The lack of a voltage causes the switch to open. There is also a transistor that is normally closed and only opens when a voltage is present - that's a NOT gate. All the other gates (AND, OR, etc.) are built up of multiple transistors. I appologise if this answer is too simplistic but without seeing what you are studying I started near the bottom. You can revise your question to narrow down the area that's confusing and we'll see if we can address that more directly.

Also, it's vital to understand that there are two types (NPN and PNP) that behave differently. Straighten out for yourself the difference between the two and that could help a lot.

Although they look simple, and are the basic building block for just about all electronic circuits, the theory and use of transistors can get quite complex.
However, once you get the hang of a few basic rules you can forget about the finer points for most circuits.
I would advise grabbing "The Art of Electronics" (quite old but a classic) and working your way slowly through the first few chapters, which are devoted to transistor theory, different types of transistors and their applications. It is very well written and gives many many good examples.
There is loads of stuff on the web, be aware that along with the good stuff there is plenty of not so good stuff. When you are starting out it's good to have confidence that you can trust what you are reading.
Of the material online All About Circuits looks quite good from the small amount I have seen.
I would recommend you get yourself a few good books though, a breadboard/stripboard, some NPN/PNP transistors and start experimenting.
SPICE can be used to simulate circuits, LTSpice is a good free tool. Careful of relying to much on it though, try to work the theory out yourself and actually build the circuits.

• All About Circuits uses electron current instead of conventional current, though, which is not a good thing for beginners. – endolith Jun 24 '14 at 13:11

I assume you not trying to learn the digital logic concepts and transistor circuits at the same time. Once you have learned each separately, it is most helpful to know that a digital output of '0' or '1' is achieved by two transistors acting in a coordinated manner such as when one is "on" the other is "off". This allows the output to be "driven by" the 5V supply when the top transistor is turned "on" while the bottom transistor is "off", or allows the output to "pulled" to ground by the bottom transistor in the opposite case. The more complex part of the circuit is needed to make sure the output transistors are turned on and off as fast as possible without overlapping their "on" or "off" times.

If you have access to some electronic parts and basic test equipment I would recommend building the '04 circuit on page four of this data sheet http://www.ti.com/lit/ds/symlink/sn74ls04.pdf. Here is a further explanation based on the '04 circuit from the above page.

The single transistor in the middle of the circuit feeding the two transistor output stage is used to make sure that the two output transistors are always turned on or off opposite of each other. When the middle transistor is "off", the bottom output transistor is turned "off", while the top one is turned "on" resulting in a logic '1' output. The opposite occurs when the middle transistor is "on", but it is a little harder to see why. Essentially, when the middle transistor is "on", both of the output transistors bases are connected together and are at a level high enough to turn on the bottom transistor, but not high enough to turn on the top one, due to the extra voltage drops in the output diode and the bottom transistor. The output is then at a logic '0'.

The trickiest part of the circuit is the input transistor which you described as "Sometimes the emitter is used as input". In this case, if nothing is connected to the input (or if 5V is applied to the input), the input transistor will be "off" and the entire input transistor node will be at the VCC (5v) level, causing the middle transistor to turn "on", the top transistor to turn "off", and the bottom transistor to turn "on" resulting in the output having a low impedance path to ground or logic level '0'.

If the input is connected to ground, the input transistor will turn "on" because the current through the 4k resistor connected to it's base. This pulls the base of the middle transistor to ground, causing the middle transistor to turn "off", the top transistor to turn "on", and the bottom transistor to turn "off" resulting in the output having a low impedance path to VCC or logic level '1'.