How transistors work (and why we need them)

Every electronic device has thousands of transistors. What do they do, and why they matter?

You may have heard that all modern digital electronics are based on transistor technology. If you ever wondered what transistor is and why it is so important, read the following article by Practicum by Yandex, and you’ll see. It’s magically simple.

What is it?

A transistor is a device that works like an electric switch. You can switch it on, and electricity will flow freely through a transistor if it’s switched off, no electricity flows. That’s it, plain and simple.

The only catch is you open and close a transistor with some other electricity. Here is a simplified scheme:

Emitter is where the power comes from. The collector is where electricity tries to go. And the transistor is a kind of a door that is locked. But when you send some power through base (it’s the pin in the middle), the door unlocks, and electricity can come through.

But what is it physically?

This question is hard to answer because a modern transistor is so small you can’t see it with your naked eye. It’s a speck of dust etched onto a silicon plate. A single transistor is so tiny, and light waves get tangled up inside those transistor arrays, which gives you this beautiful rainbowy light effect when you look at modern transistors:

Each segment of this plate contains hundreds of millions, possibly billions of transistors. They are so densely packed, light waves get tangled up between them and produce this rainbow-like color

But you still can see transistors like this:

This is a transistor cased in plastic, used in educational projects, and DIY electronics. It works the same way as any other transistor; only it’s enormous compared to modern-day nanometer transistors.

How it works

The transistor itself has a straightforward principle: let electricity pass when open; block electricity when closed. That’s it. Open and close, one and zero. This isn’t particularly useful in itself.

The key here is how you connect those transistors. What if one transistor-controlled how another transistor worked? What if one transistor’s collector is another transistor’s base? What if two transistors shared the same collector?

We will dedicate a separate article to this subject, but for now, it will suffice to say this:

1. Transistors allow us to calculate basic logic if they are connected in a certain way. For example, using only one transistor, you can make a device that inverts your signal: you tell it “on”, it outputs “off”, and vice versa.
2. Four logical operations can be made with interconnected transistors. Each operation is represented by a certain arrangement and connections of transistors.
3. If you combine these arrangements, you can create a machine that helps you make a sum of two numbers — a summing device, which is essentially a computer.
4. Have enough summing devices, and you have yourself a proper computer.

How transistors came to be

Before transistors, engineers and scientists tried to build computing machines, and many succeeded. They used mechanical parts, like gears, rotors, and springs. But their mechanical calculators were pretty large, not very capable and rather expensive.

In the times of WWII, there was a Turing Machine, boldly depicted in The Imitation Game (catch it around 1:24):

The Turing Machine, aside from being based on a breakthrough theoretical framework, used mechanical components to control the signal. They could be slow and faulty.

Next came the vacuum tubes or lamps. Yes, lamps.

Lamps allowed the signal to pass through them when there was a controlling signal present. But, of course, lamps were fragile, prone to overheating, and not very fast.

And finally, scientists discovered how to use semiconductors in controlling the current, and along came transistors. They were relatively fast, reliable, hard to break, and easy to maintain, so everyone stuck with transistors.

From then on, transistors became smaller and smaller, faster and faster, and here we are: there are 8,5 billion transistors in an Apple A13 processor that you can fit on your thumb. And yes, it is used to turn someone’s face into a three-dimensional singing poop.

Why they matter

Transistors make up logical circuits. Arrays of logical circuits make up computational circuits. Arrays of computational circuits make up processors and controllers. Those guys control the rest of the world: a microwave, a coffee maker, an electric kettle, a TV, a smartphone, a laptop, a Pentagon supercomputer, an Apollo 11, a Tesla — everything is filled with densely packed and fancily connected transistors that turn each other on and off fast.

Next time we’ll look at all those arrangements, and you’ll see how simple it is to achieve transistor supremacy. The only problem is, quantum supremacy is coming, and we’ll talk about that too.