Head Sliders, Arms and ActuatorThe hard disk platters are accessed for read and write operations using theread/write heads mounted on the top and bottom surfaces of each platter. Obviously, the read/write heads don't just float in space; they must be held in an exact position relative to the surfaces they are reading, and furthermore, they must be moved from track to track to allow access to the entire surface of the disk. The heads are mounted onto a structure that facilitates this process. Often called the head assembly or actuator assembly (or even the head-actuator assembly), it is comprised of several different parts.
The heads themselves are mounted on head sliders. The sliders are suspended over the surface of the disk at the ends of the head arms. The head arms are all mechanically fused into a single structure that is moved around the surface of the disk by the actuator. (Sort of like "the leg bone's connected to the knee bone", isn't it? :^) ) It would be an understatement to say that these components are neglected; heck, most people don't even know they exist! Yet they play an important role in the function and performance of the drive. In particular, advances in slider, arm and actuator design are critical to improving the seek time of a hard disk; the impact that the actuator has on performance is discussed in this section.
This section discusses the sliders, arms and actuator of the modern , as well as explaining the operation of disk drive servo mechanisms and related .
Annotated illustration of a typical PC actuator assembly, showing the major
components. The platters have been removed from the drive to provide a better
view of the actuator arms and heads. There are four sliders but only
one of each pair is visible. The spindle motor is visible at the top right.
Head Sliders :
Hard disk read/write heads are too small to be used without attaching them to a larger unit. This is especially true of modern hard disk heads. Each hard disk head is therefore mounted to a special device called a head slider or just sliderfor short. The function of the slider is to physically support the head and hold it in the correct position relative to the platter as the head floats over its surface.
Sliders are given a special shape to allow them to ride precisely over the platter. Usually they are shaped somewhat like a sled; there are two rails or runners on the outside that support the slider at the correct flying height over the surface of the disk, and in the middle the read/write head itself is mounted, possibly on another rail.
As hard disk read/write heads have been shrinking in size, so have the sliders that carry them. The main advantage of using small sliders is that it reduces the weight that must be yanked around the surface of the platters, improving both positioning speed and accuracy. Smaller sliders also have less surface area to potentially contact the surface of the disk.
The hard disk industry has given names to various generations of slider. When the design was first reduced to 50% of the size of the first hard disk sliders, someone decided to call it the nano-slider, where "nano" is the prefix denoting "one billionth". Of course the name is silly, since the sliders have not shrunk by even a factor of 10. The newest sliders have been shrunk from the "nano" size by about another 50% and are being called pico-sliders, which in this author's opinion is an equally silly name, and for the same reason. :^)
A graphic illustration of what approximately 15 years' worth of
technological evolution has done to hard disk head sliders.
At left, a slider from a 40 MB 5.25" ferrite-head drive;
at right, the slider from a 3.2 GB, 3.5" MR-head drive.
Each slider is mounted onto a head arm to allow it to be moved over the surface of the platter to which it is mated.
Head Arms :
The head arms are thin pieces of metal, usually triangular in shape onto which the head sliders (carrying the read/write heads) are mounted. In a way, the idea here is similar to how the arm of a phonograph is used to move the stylus from the outside of a record to the inside (although of course the similarity ends there). There is one arm per read/write head, and all of them are lined up and mounted to the head actuator to form a single unit. This means that when the actuator moves, all of the heads move together in a synchronized fashion. Heads cannot be individually sent to different track numbers.
The arms themselves are made of a lightweight, thin material, to allow them to be moved rapidly from the inner to outer parts of the drive. Newer designs have replaced solid arms with structural shapes in order to reduce weight and. This is the same technique used to reduce weight in the construction of airplane wings, for example. Newer drives achieve faster seek times in part by using faster and smarter actuators and lighter, more rigid head arms, allowing the time to switch between tracks to be reduced.
A recent trend in the hard disk industry has been the reduction in the number of platters in various drive families. Even some "flagship" drives in various families now only have three or even two platters, where four or five was commonplace a year or so ago. One reason for this trend is that having a large number of head arms makes it difficult to make the drive with high enough precision to permit very fast positioning (on random seeks). This is due to increased weight in the actuator assembly from the extra arms, and also problems aligning all the heads. So in essence, this is a tradeoff that some drive manufacturers are making to improve performance at the expense of capacity. With drive densities now at 20 GB per platter and bound to increase, this is an acceptable design decision for most buyers.
Head Actuator :
The actuator is the device used to position the head arms to different tracks on the surface of the platter (actually, to different cylinders, since all head arms are moved as a synchronous unit, so each arm moves to the same track number of its respective surface). The actuator is a very important part of the hard disk, because changing from track to track is the only operation on the hard disk that requires active movement: changing heads is an electronic function, and changing sectors involves waiting for the right sector number to spin around and come under the head (passive movement). Changing tracks means the heads must be shifted, and so making sure this movement can be done quickly and accurately is of paramount importance. This is especially so because physical motion is so slow compared to anything electronic--typically a factor of 1,000 times slower or more.
Head actuators come in two general varieties:
- Stepper Motors: Originally, hard disk drives used a stepper motor to control the movement of the heads over the surface of the platters. A regular motor turns in a rotary fashion continuously; it can stop at any point in its rotation as it spins around, kind of like the second hand on a wind-up wristwatch. A stepper motor can only stop at predefined "steps" as it turns around, much the way the second hand turns on an electronic, quartz wristwatch. A using a stepper motor for an actuator attaches the arms to the motor, and each time the motor steps one position clockwise or counterclockwise, the arms move in or out one position. Each position defines a track on the surface of the disk. Stepper motors are also commonly used for both turning the spindle and positioning the head on . If you have a floppy drive, find one of its motors and turn it slowly with your hand; you will feel the discrete step-wise nature of its motion.
A stepper motor actuator. The motor moves in steps, which you can feel if
you move the motor shaft by hand. The shaft has two thin strips of metal
wrapped around it, which are connected to a pivot that is rigidly attached
to the actuator arms. As the motor shaft turns, one half of this "split band"
coils onto the shaft and the other half uncoils. When the motor turns in the
opposite direction the process reverses. As this occurs the pivot moves
and in doing so, moves the actuator arms and the hard disk heads.
- Voice Coils: The actuator in a modern hard disk uses a device called avoice coil to move the head arms in and out over the surface of the platters, and a closed-loop feedback called a servo system to dynamically position the heads directly over the data tracks. The voice coil works using electromagnetic attraction and repulsion. A coil is wrapped around a metal protrusion on the end of the set of head arms. This is mounted within an assembly containing a strong permanent magnet. When current is fed to the coil, an electromagnetic field is generated that causes the heads to move in one direction or the other based on attraction or repulsion relative to the permanent magnet. By controlling the current, the heads can be told to move in or out much more precisely than using a stepper motor. The name "voice coil" comes from the resemblance of this to that used to drive audio speakers, which are also basically electromagnets. All PC hard disk voice coil actuators are rotary, meaning that the actuator changes position by rotating on an axis.
The primary distinction between the two designs is that the stepper motor is anabsolute positioning system, while the voice coil is a relative positioning system. Commands given to a stepper motor actuator are generally of the form "Go in this direction to position A, where you'll find item B". Commands to a voice coil actuator are of the form "Go in this direction until you find item B". Consider this analogy. In your backyard you have buried a "secret treasure" and want to tell a friend where to find it. When you buried it, you walked down a path 139 paces to the fourth oak tree, and buried it at the edge of the path. The stepper motor analog would be to tell your friend to walk 139 paces down the path, and start digging. The voice coil analog would be to tell him to look for the fourth oak tree and dig there. Obviously, using the "139 paces" method, your friend has a problem: his paces aren't likely to be the same length as yours. In fact, even if you yourself walked out 139 paces twice, you'd probably end up in very different spots, since a "pace" isn't an accurate or repeatable measure. On the other hand, the fourth oak tree will always be the fourth oak tree (barring disastrous chain-saw activity :^) ).
A partially-disassembled voice coil actuator. The magnet assembly has
been unscrewed from its mounting and pulled to the left to expose the
coil. The magnet assembly consists of two metal plates (top one easily
visible above, and part of the bottom one visible.) The magnet itself is
mounted on the underside of the top plate, and spacers used between
the plates to create the gap for the coil assembly. Being non-ferrous the
coil moves freely between the plates, rotating the actuator on its axis
as its magnetic polarity is changed. (Incidentally, the magnet is strong
enough that after removing the spacers between the plates, the bottom plate
got "stuck" on the magnet and required considerable effort to remove!)
Now hard disks of course don't have to use inaccurate measures like "paces", and it's always the same stepper motor accessing the disk, not a "friend", so why is saying "track #139" a big problem? For starters, motors change their characteristics over time, and after a year or two position #139 might not be where it was when the drive was first formatted. However, they have an even more serious problem: disk components (the platters and the head arms themselves especially) expand and contract with heat. Even if a stepper motor was perfect, it could not properly account for the fact that the disks are changing in size, and therefore, the tracks are literally moving around. If you consider our backyard analogy and think about what it would be like if the oak tree moved a few feet closer to or further from the house based on the day's temperature, you start to realize how inadequate absolute positioning of this form can be.
A stepper motor has no way to compensate for expansion or contraction of the disk: all it can do is go to where "track #139" is supposed to be, and hope it finds it there! If it doesn't find it because the motor and the disk have become out of sync, errors and data loss result. This is why older disks were so sensitive to temperature, and normally had to be low-level formattedperiodically to make sure the tracks lined up with the heads properly. This is also why many drives would fail when first powered up after a weekend, but would work properly after the drive had been allowed to warm up.
The shortcomings of stepper motors were unfortunate but acceptable with old hard disks, because of their relatively low track density. To compensate, tracks could be written fairly wide so that the head would find them even if it was a bit misaligned. The first PC hard disks in 1982 had a track density of only two or three hundred tracks per inch (TPI). Even In 1986, the year Conner Peripherals introduced the first voice coil PC hard disk, density had increased to only about 1,000 TPI. Stepper motors are still used to drive floppy disks, for example, because the accuracy demands for floppies are much lower: a 1.44 MB floppy disk has a track density of 135 tracks per inch. In contrast, today's hard disks have densities as high as 30,000 tracks per inch. With data packed this densely, tracks are extremely thin, and a stepper motor lacks the accuracy and stability necessary for proper operation.
All modern hard disks use voice coil actuators. The voice coil actuator is not only far more adaptable and insensitive to thermal issues, it is much faster and more reliable than a stepper motor. The actuator's positioning is dynamic and is based on feedback from examining the actual position of the tracks. This closed-loop feedback system is also sometimes called a servo motor or servo positioning system and is commonly used in thousands of different where precise positioning is important. There are several different ways that the servo positioning system is implemented in PCs; the servo's operation is discussed in its own section.