Wirebook › Bonus lesson
How does a TV remote work when nothing comes out of it?
You press a button, and the TV on the other side of the room changes channel. Nothing seems to come out of the remote, no beam, no sound. But something does: a tiny lamp at the front flashes infrared light, a colour just past red that our eyes can't see. The flashes are a code, like Morse code done at lightning speed, and the TV has a little eye that reads it. Here you can press the buttons, see the hidden flashes, and read a code yourself.
Light you can't see
Light comes in waves, and our eyes only notice some sizes of wave: violet, blue, green, yellow, red. Waves a little longer than red are called infrared. The lamp in the front of a remote is a small LED that glows in infrared, so to us it looks dark even while it's working. Many phone cameras can see a bit of infrared, which is why the remote's tip flashes purple-white on a phone screen. That's a handy test: if you see the flash on your phone but the TV does nothing, the remote is fine and the problem is somewhere else. (Some phones' main cameras filter infrared out, so try the selfie camera.)
A secret flicker so the sun can't fool it
Rooms are full of light, and sunshine and warm things give off infrared too. So the remote doesn't just switch its lamp on and off. Every flash is really a burst of very fast flickering, about 38,000 times a second. The TV's little eye has a chip tuned to that one flicker speed, a bit like a friend who only answers to one special whistle. Steady light from the sun or a lamp doesn't flicker like that, so it ignores it. Try the sunny window with and without the flicker trick.
Long gaps and short gaps make a code
A common remote code goes like this. First a long wake-up flash (9 thousandths of a second) and a pause, so the TV knows a message is coming. Then 32 short bursts. What matters is the gap after each burst: a short gap means 0 and a gap three times longer means 1. The first 8 bits say which brand or device the message is for (the address), and the next 8 are the same address with every bit flipped. Then come 8 bits for the button (the command) and the command flipped again. The flipped copies are a check: if a speck of sunlight messed up one bit, the copy won't match and the TV ignores the message rather than doing the wrong thing. If you hold a button down, the remote sends a short "still pressed" signal about ten times a second, which is how the volume keeps going up.
Why doesn't my remote change the neighbour's TV?
Each brand uses its own address at the start of the code, and sometimes its own flicker speed and gap pattern too. A TV listens to every message its eye catches but only obeys the ones with its own address, so the neighbour's remote is just noise to it. Universal remotes and phone apps that work as remotes keep a big list of brands' addresses and button codes, which is why you pick your TV's brand when you set one up.
When the remote stops working
Infrared travels in straight lines and bounces off pale walls and ceilings, which is why pointing roughly in the right direction usually works. It can't go through cushions, people or the edge of a cabinet. Tired batteries make dimmer flashes, so the remote only works close up: that's often the first sign the batteries need changing. Very bright sunlight or some energy-saving lamps right next to the TV can drown the eye. And many newer smart TV remotes don't use infrared for every button at all: they talk to the TV by Bluetooth radio, which goes through cushions but has to be paired first.
Related lessons: how face unlock uses invisible infrared dots · how Wi-Fi radio waves get around your house · how a phone recognises a song · how a game controller talks to a console.
Grown-up twin: Tooltern's help page on testing a remote with a phone camera, and fixing one that stops working.