How to store information with magnetism
The science of magnetism is complex. But if you’ve ever fooled around with a magnet and some nails, you’ll know that the technology—the science in action—is quite simple. Iron nails start off unmagnetized but, if you rub a magnet back and forth over them, you can make them magnetic so they stick to one another. Magnetism has some simple, practical uses. For example, junkyards use electromagnets (huge magnets that can be switched on and off with electricity) to pick up and move around piles of metal scrap.
Magnetism has another very important use. Suppose you need to leave a message for a friend and all you have is a magnet and an unmagnetized iron nail. Suppose the message is a very simple one: either you will see your friend later that day or not. You could arrange with your friend that you will drop a nail through their letterbox. If the nail is magnetized, it means you will see them later; if the nail is unmagnetized, you won’t. Your friend gets in from school and finds a nail on the doormat. They take it to the kitchen table and try to pick up a paperclip. If the clip attaches to the magnet, it must be magnetized—and it must mean you plan to see them later. It’s a pretty weird way to leave a message for someone, but it illustrates something very important: magnetism can be used to store information.
If your computer has a 20 gigabyte (GB) hard drive, or you have a 20 GB iPod or MP3 player, it’s a bit like a box containing 160 thousand million microscopically small iron nails, each of which can store one tiny piece of information called a bit. A bit is a binary digit—either a number zero or a number one. In computers, numbers are stored not as decimal (base-10) but as patterns of binary digits instead. For example, the decimal number 382 is stored as the binary number 101111110. Letters and other characters can also be stored as binary numbers. Thus, computers store a capital letter A as the decimal number 65 or the binary number 1000001. Suppose you want to store the number 1000001 in your computer in that big box of iron nails. You need to find a row of seven unused nails. You magnetize the first one (to store a 1), leave the next five demagnetized (to store five zeros), and magnetize the last one (to store a 1).
How a hard drive works
In your computer’s hard drive, there aren’t really any iron nails. There’s just a large shiny, circular “plate” of magnetic material called a platter, divided into billions of tiny areas. Each one of those areas can be independently magnetized (to store a 1) or demagnetized (to store a 0). Magnetism is used in computer storage because it goes on storing information even when the power is switched off. If you magnetize a nail, it stays magnetized until you demagnetize it. In much the same way, the computerized information (or data) stored in your PC hard drive or iPod stays there even when you switch the power off.
What are the parts in a hard drive?
A hard drive has only a few basic parts. There are one or more shiny silver platters where information is stored magnetically, there’s an arm mechanism that moves a tiny magnet called a read-write head back and forth over the platters to record or store information, and there’s an electronic circuit to control everything and act as a link between the hard drive and the rest of your computer.
After a hard-drive crash last year, I was left with an old drive that no longer worked. I took a peek inside, and here’s what I found…
- Actuator that moves the read-write arm. In older hard drives, the actuators were stepper motors. In most modern hard drives, voice coils are used instead. As their name suggests, these are simple electromagnets, working rather like the moving coils that make sounds in loudspeakers. They position the read-write arm more quickly, precisely, and reliably than stepper motors and are less sensitive to problems such as temperature variations.
- Read-write arm swings read-write head back and forth across platter.
- Central spindle allows platter to rotate at high speed.
- Magnetic platter stores information in binary form.
- Plug connections link hard drive to circuit board in personal computer.
- Read-write head is a tiny magnet on the end of the read-write arm.
- Circuit board on underside controls the flow of data to and from the platter.
- Flexible connector carries data from circuit board to read-write head and platter.
- Small spindle allows read-write arm to swing across platter.
The platters are the most important parts of a hard drive. As the name suggests, they are disks made from a hard material such as glass, ceramic, or aluminum, which is coated with a thin layer of metal that can be magnetized or demagnetized. A small hard drive typically has only one platter, but each side of it has a magnetic coating. Bigger drives have a series of platters stacked on a central spindle, with a small gap in between them. The platters rotate at up to 10,000 revolutions per minute (rpm) so the read-write heads can access any part of them.
There are two read-write heads for each platter, one to read the top surface and one to read the bottom, so a hard drive that has five platters (say) would need ten separate read-write heads. The read-write heads are mounted on an electrically controlled arm that moves from the center of the drive to the outer edge and back again. To reduce wear and tear, they don’t actually touch the platter: there’s a layer of fluid or air between the head and the platter surface.
Reading and writing data
The most important thing about memory is not being able to store information but being able to find it later. Imagine storing a magnetized iron nail in a pile of 1.6 million million identical nails and you’ll have some idea how much trouble your computer would get into if it didn’t use a very methodical way of filing its information.
When your computer stores data on its hard drive, it doesn’t just throw magnetized nails into a box, all jumbled up together. The data is stored in a very orderly pattern on each platter. Bits of data are arranged in concentric, circular paths called tracks. Each track is broken up into smaller areas called sectors. Part of the hard drive stores a map of sectors that have already been used up and others that are still free. (In Windows, this map is called the File Allocation Table or FAT.) When the computer wants to store new information, it takes a look at the map to find some free sectors. Then it instructs the read-write head to move across the platter to exactly the right location and store the data there. To read information, the same process runs in reverse.
How does an electronic computer manipulate all the mechanical nitty gritty in a hard drive? There is an interface (a connecting piece of equipment) between them called a controller. This is a small circuit that operates the actuators, selects specific tracks for reading and writing, and converts parallel streams of data going from the computer into serial streams of data being written to the disk (and vice versa). Controllers are either built into the disk drive’s own circuit board or part of the computer’s main board (motherboard).
With so much information stored in such a tiny amount of space, a hard drive is a remarkable piece of engineering. That brings benefits (such as being able to store 500 CDs on your iPod)—but drawbacks too. One of them is that hard drives can go wrong if they get dirt or dust inside them. A tiny piece of dust can make the read-write head bounce up and down, crashing into the platter and damaging its magnetic material. This is known as a disk crash (or head crash) and it can (though it doesn’t always) cause the loss of all the information on a hard drive. A disk crash usually occurs out of the blue, without any warning. That’s why you should always keep backup copies of your important documents and files, either on another hard drive, on a compact disc (CD) or DVD, or on a flash memory stick.