How A clock and watches Works

You may not believe in astrology, but there's no question the planets rule our lives. We get up when the Sun rises (or some time after) and go to bed when it sets. We have a calendar based on days, months, and years—periods of time that relate to how the Moon and Earth move around the Sun in the sky.

For most of history, people found this kind of "astronomical timekeeping" good enough for their needs. But as the world became ever more frantic and sophisticated, people needed to keep track of hours, minutes, and seconds as well as days, months, and years. That meant we needed accurate ways of keeping time. Pendulum clocks and mechanical watches used to be the best way of doing this. Today, many people use quartz clocks and watches instead—but what are they and how do they work?

How ordinary clocks work

We all know that a clock keeps time, but have you ever stopped to think about how it does so? Probably the simplest clock you could make is a speaking clock. If you count seconds by repeating a phrase that takes exactly one second to say (Like "elephant one", "elephant two", "elephant three"...), you'll find you can keep time pretty accurately. Try it out. Say your elephants from one to sixty and see how well you keep time over a minute, compared to your watch.

Not bad, eh? The trouble is, most of us have better things to do all day than say "elephant." That's why people invented clocks. Some of the earliest clocks used swinging pendulums to keep time. A pendulum is a long rod or a weight on a string that swings back and forth. In 1583, the Italian physicist Galileo Galilei (1564–1642) discovered that a pendulum of a certain length always takes the same time to swing back and forth, no matter how heavy it is or how big a swing it makes. He figured this out by watching a huge lamp swinging on a chain from the ceiling of Pisa Cathedral in Italy, and using his pulse to time it as it moved back and forth. In a clock, the pendulum's job is to regulate the speed of the gears (interlocking wheels with teeth cut into their edges). The gears count the number of seconds that pass and convert them into minutes and hours, displayed on the hands that sweep round the clockface. To put it another way: the gears in a pendulum clock are really just counting elephants.

You can make a pendulum clock by tying a weight to a piece of string. If the string is about 25cm (10 inches) long, the pendulum will swing back and forth roughly once each second. Shorter strings will swing faster and longer strings slower. The trouble with a clock like this is that the pendulum will keep stopping. Air resistance and friction will soon use up its energy and bring it to a halt. That's why pendulum clocks have springs in them. Once a day or so, you wind up a spring inside the clock to store up potential energy to keep the pendulum moving for the next 24 hours. As the spring uncoils, it powers the gears inside the clock. Through a see-saw mechanism called an escapement, the pendulum forces the gears to turn at a precise rate—and this is how the gears keep time. A pocket watch is obviously too small to have a pendulum inside it, so it uses a different mechanism. Instead of a pendulum, it has a balance wheel that turns first one way and then the other, controlled by a much smaller escapement than the one in a pendulum clock.

Inside a quartz clock

Artwork showing how a quartz watch works

In theory, it works like this:

Battery provides current to microchip circuit
Microchip circuit makes quartz crystal (precisely cut and shaped like a tuning fork) oscillate (vibrate) 32768 times per second.
Microchip circuit detects the crystal's oscillations and turns them into regular electric pulses, one per second.
Electric pulses drive miniature electric stepping motor. This converts electrical energy into mechanical power.
Electric stepping motor turns gears.
Gears sweep hands around the clockface to keep time.

In practice...

And this is what the inside of a quartz watch looks like in reality. Don't, under any circumstances, take yours apart if you ever want it to work again. You cannot see all these parts just by taking the back off a watch. The watch shown here came free with a packet of cornflakes (seriously!) and it was broken before I opened it up. But it was even more broken afterwards...

Battery.
Electric stepping motor.
Microchip.
Circuit connects microchip to other components.
Quartz crystal oscillator.
Crown screw for setting time.
Gears turn hour, minute, and second hands at different speeds.
Tiny central shaft holds hands in place.

Why do quartz watches gain or lose time at all?

If quartz is so amazing, you might be wondering why a quartz watch doesn't keep time with absolutely accuracy forever. Why does it still gain or lose seconds here and there? The answer is that the quartz vibrates at a slightly different frequency at different temperatures and pressures so its timekeeping ability is affected to a tiny degree by the warming, cooling, ever-changing world around us. In theory, if you keep a watch on your wrist all the time (which is at more or less constant temperature), it will keep time better than if you take it on and off (causing quite a dramatic temperature change each time). But even if the quartz crystal could vibrate at a perfectly constant frequency, the way it's mounted in its circuit, tiny imperfections in the gearing, friction, and so on can also introduce minute errors in timekeeping. All these effects are enough to introduce an inaccuracy of up to a second a day in typical quartz clocks and watches (bear in mind that a second lost one day may be compensated by a second gained the next day, so the overall accuracy may be as good as a few seconds a month).

But how does the quartz crystal bit actually work?

You might find that enough of an explanation and, if so, you can stop reading now. What follows is a more detailed discussion of how the quartz crystal oscillator actually works for those who want to a bit more depth. I should warn you that unless you have a degree in electronic engineering, quartz crystal circuits get very complex very quickly. I'm going to give you a very brief, simplified version of what's happening and some pointers for further reading so you can dig deeper if you care to.

The key thing to remember about quartz is that it's piezoelectric: it will vibrate when you put electricity into it, or it will give out electricity when you vibrate it. A quartz crystal oscillator uses piezoelectricity in both ways—at the same time!

The way I've drawn my diagram up above makes it look like the quartz crystal is separate from the microchip circuit but, in reality, the crystal is an intimate part of that circuit, wired into it by two electrodes. You can see them clearly in the large photo of the watch's insides and in the photo of the oscillator itself: they're the two little silver-colored legs poking out from the cylindrical metal case. In effect, the quartz crystal oscillator is just another component wired into the microchip circuit, just like a resistor or a capacitor.

I say "circuit" but it's simplest to think of the oscillator as being part of two separate circuits, both of which are on the same microchip. The first circuit (we'll call it the input) stimulates the quartz crystal with bursts of electricity. Feeding electricity into quartz makes it vibrate (or, if you prefer, oscillate or resonate) through what's sometimes called the reverse piezoelectric effect (where electricity produces vibrations). The oscillator is set up so the quartz vibrates exactly 32768 times a second. But now remember the normal piezoelectric effect: when a piece of quartz vibrates, it generates an electrical voltage. The second circuit on the microchip detects this "output voltage" (fluctuating 32768 times a second) and divides its frequency to produce once-a-second pulses that drive the motor powering the gears. In a watch with a digital display, instead of using gears, a chip repeatedly divides the oscillator frequency to drive the hours, minutes, and seconds segments (as shown in the artwork below).

Quartz oscillator, binary divider, counters, and display drivers in a typical Timex digital watch from US Patent 3,863,436.

Artwork: How a quartz oscillator drives a digital watch with an hours and minutes display and a flashing colon between them ("12:32") to indicate the passing seconds. The oscillator (yellow) vibrates 32,768 times a second. A binary divider (blue, left) divides this 15 times to create a 1 Hz (one per second) pulse that drives the flashing colon. The 1Hz signal from the divider is itself divided by 60 to make minutes and another 12 to make hours. These signals operate a series of drivers (red) that power the segments in the digital display. Artwork from US Patent 3,863,436: Solid state quartz watch by Jack Schwarzschild and Raymond Boxberger, Timex. February 4, 1975, courtesy US Patent and Trademark Office.

In one early form of quartz oscillator, the quartz crystal had two sets of electrodes mounted on it. The first set was connected to the input circuit and fed electricity into the crystal to make it vibrate. When the crystal vibrated, it generated a piezoelectric voltage. That was detected by the second set of electrodes (stuck to a different part of the same crystal) and fed to the output circuit. When quartz technology was miniaturized for use in compact wristwatches, it became clear that smaller oscillators were needed and there wasn't room for two pairs of electrodes. That's why modern oscillators use a single pair of electrodes both to stimulate the crystal with energy and detect its vibrations.

That's as much as I'm going to tell you. If you want to find out more, you might like to take a look at the following sources. Be warned that they are complex and hard to follow unless you have some knowledge of electronic engineering.

Search This Blog