How Do Hard Drives Work?
It doesn’t matter what you call it; hard drive, hard disk, hard disk drive, HDD, it’s all the same. There are very few components within a computer that are as complicated as a hard drive. It is everything when it comes to our computers. Without them, our computers would be lifeless and empty. But hard drives are the soul of our modern computers. Hard drives store our lives, our memories, our businesses…everything. Hard drives make up the majority of our data recovery jobs. So how does a hard drive work? You may speak with a lot of data recovery companies, but very few will actually be able to accurately describe what it is a hard disk does, and how it works.
Hard Drive Basics
Our data is saved to a hard drive in it’s most basic form, known as binary digits. Binary digits are also known as bits. A bit is either “on” or “off”. We see this in digital format as 1’s and 0’s. But how do those 1’s and 0’s get written to the hard drive?
In short, the bits are written to the hard disk by a head. The head is actually referred to as a read-write head, because it is used to either write data or read data. The head is an electromagnet. Electromagnets consist of a piece of metal, surrounded by a coil of wire. An electric current passes through the coil of wire and creates a magnetic field. This magnetic field then magnetizes whatever bits that need to be 1’s and bypasses the bits that need to be 0’s. So in summary, you have the read-write heads either reading the disk by simply measuring the magnetic polarization (deciphering whether a bit is a 1 or a 0) or they are writing data by altering the magnetic polarization in specific areas of the disk.
Key Components of a Hard Drive
It’s almost scary to think of how precarious hard drives are. Their design and operation is so precise, but think of the environment in which we use them. While in most instances they might just be sitting there in your desktop computer or a server humming along with each passing day, they are also found in many portable devices. They are carried, dropped, knocked around sometimes, and most keep on working. Yet when you really think about how intricate…how fragile they are, you’d probably treat it as a new born baby.
Since the beginning hard drives have been made up of roughly the same components. The designs may have improved over the years, but you still have a core group of parts that function pretty much as they always have.
- Platter – The actual “disk”. The part of your drive that actually stores the data.
- Read-Write Heads – Pretty much self-explanatory. The read-write heads, read and write data to the platter
- Actuator – Controls the movement of read-write heads as they navigate their way over the platter
- Spindle – The hub on which the platters are mounted
- PCB (Printed Circuit Board) – Where you attach your power and interface cables. The PCB also stores data specific to your hard drive, offers some voltage surge protection and regulates other functions within the drive.
When you look at a hard drive’s internal components, it’s similar to a record player. You have the platters, which are like the record…and you have the read-write heads, which are like the needle. The heads read and write data to the platter surface.
The platters, which are the rotating disks inside the hard drive, are rigid and typically manufactured from aluminum, ceramic, or glass material. It’s because the disks are rigid that hard drives are called “hard” drives, as opposed to floppy drives that used flexible disks for data storage. Data can be stored on both sides of a platter, and in most cases a hard drive will have multiple platters affixed to a single spindle. Each platter surface has it’s own dedicated read-write head. So a hard drive with 3 platters, will most likely have 6 heads. This isn’t always the case, but more often than not, it is.
The read-write heads are mounted at the very tip of a single actuator arm. The very tip of the head assembly consists of the read-write heads, and what’s referred to as a slider. Before we get into the function of the slider, it’s important to notate an important fact about hard drives. The heads do not touch the platters at all during normal operation. They actually float on a cushion of air, which is referred to as, an air bearing. As the platters rotate beneath the heads, they create positive air pressure, and the slider acts almost like the wing on an airplane. The air pressure acting against the slider, lifts the heads and they float just above the platter surface. In modern hard drives, the amount of space between the head and rotating platter at normal operating speed is typically less than 5 nanometers…this gap is also referred to as the flying height.
This actuator arm moves back and forth allowing the heads to sweep across the platter surface to their required location. The actuator arm, and therefore the movement of the heads, is controlled by an actuator. In older hard drives, this was simply a stepper motor, but in most drives built within the last 10+ years, a voice coil actuator is used. This voice coil, is surrounded by stationary magnetic surfaces above and below the base of the actuator arm. A voice coil actuator is essentially an electromagnet. The amount of current passing through the voice coil is what controls the direction in which the actuator arm rotates in relation to the magnets that surround it.
Reading and Writing Data To The Platters
Hard drive read-write heads write data to the platters by altering the polarization of the top layer of the platter surface. This top layer consists of a very thin ferromagnetic coating. Just saying the coating is thin, doesn’t really do it justice. The coating is only about 10-20 nanometers thick. How thin is that? Well, if you want to compare, a piece of paper is usually 70,000 to 180,000 nanometers thick. Now this is where things really get mind boggling and down right scary when you think about it, and here’s why…
- The thin ferromagnetic coating is extremely thin, and susceptible to extensive damage even with the most minimal of contact.
- The coating contains your data…it holds the magnetic switching 1’s and 0’s that make up your pictures, your documents, your life in many cases.
- The read-write heads float on a cushion of air, with a flying height in newer drives of usually less than 5 nanometers above the surface of the platter.
- In case you missed that…the heads float less than 5 nanometers above the platter surface!
- For reference to point 3 and 4, a single strand of DNA is 2.5 nanometers wide…a single bacterium is 2500 nanometers long…a strand of hair is 80,000 nanometers wide
- The air gap between the heads and the platter is so small, a single strand of DNA would have a hard time passing through it.
Are you starting to get a feel for how small these clearances are? When we talk about repairing hard drives and the need for clean rooms, this is why it is so important. A class-100 clean room is sufficient for working with hard drives, but anything less than that can cause significant problems. When we receive hard drives that have been opened, in attempts to recover the data themselves, the contamination of the platter is usually extensive. Most people don’t even realize how catastrophic a single finger print can be. Just to reiterate, the air gap between the heads and platters is typically less than 5 nanometers in newer hard drives. Did you know that a single fingerprint has a thickness of over 12,000 nanometers?
Just as an aside, since while we are on the subject of do it yourself data recovery attempts…let’s talk briefly about freezing a hard drive to recover the data. It’s an old myth that is an oft recommended as a solution. What you may not know is frost can and does form on the platter surface. These tiny microscopic ice crystals are still in excess of 30,000 nanometers in height. Imagine the air gap between the heads and the platter is your typical highway overpass. Now imagine trying to fit Mount Everest under that overpass. That is what it’s like when you freeze your drive. We don’t mention that to scare you into using our services, we do it to explain why it’s such a bad idea to do something that is recommended so often. And freezing a hard drive has never ever repaired a damaged set of heads. A drive that works after being put in a freezer had something entirely different wrong with it.
If the above points don’t give you a good sense of the complexity of a hard drive, let’s put it into a real world scale. As the heads are floating over the platter surface, they are able to sweep back and forth with precise movements at blistering speeds to either read or write data to specific areas on the platter. If you were to scale this up, imagine a fighter jet traveling in excess of Mach 5 (nearly 4,000MPH) at less than 1-inch off the ground, and being able to stop on any given blade of grass. That is how precise hard drives are.
Data is written to the platter through the transmission of an electromagnetic flux. This is delivered via the read-write heads. As the heads pass over the rotating platter surface, the polarization of the magnetic coating is changed due to the flux that is passed through the read-write head. In a way it is zapping a magnetic pulse at precise locations. Data is read as the head passes over the rotating platter. Differences in magnetism are detected by the head and that generates a current, which is then interpreted as a binary 1 or 0.
Large Capacity Drives
Not long ago it was thought that hard drive capacities would be forever limited to 3TB due to physical limitations. This barrier has been effectively obliterated with drives well in excess of 3TB. The problem with capacity is data density. The more data you can squeeze onto a single platter surface, the bigger the overall drive will be. A few years back Seagate announced that they expected to release 60TB hard drives by 2019. As of this writing, HGST a Western Digital company, launched 10TB helium filled hard drives.
So how are they achieving such large capacities, when only a short time ago 3TB was the limit? There has been ongoing development in the way data is written to and stored on the platters. The physical size of the platters are not expected to change, but the amount of data they can squeeze onto the platter (often referred to as data density) is getting tighter and tighter.
Recording methods are also changing. Over the last few years the way in which data is written to hard drives platters has changed. It used to be with older style hard drives, the data was written to the platter Longitudinally, but starting in 2005 that began to switch over to Perpendicular recording. This allowed more bits / more data per square inch of platter.
Now there are even bigger breakthroughs in how data is written to platters. Drive manufacturers are starting to use Heat Assisted Magnetic Recording – HAMR, which uses lasers to heat the surface before data is written. These storage methods will use a iron platinum alloy, which is extremely stable. HAMR technology has enable Seagate to achieve a data density of 1TB per square-inch.
Hard Drive Teardown
This is a good video showing the breakdown of a hard drive and how it operates.
Written summary below…
A home computer is a powerful tool, but it must store data reliably to work well, otherwise its kind of pointless isn’t it. Let’s look inside and see how it stores data. Look at that. It’s marvelous. It’s an ordinary hard drive, but its details, of course, are extraordinary. Now, I’m sure you know the essence of a hard drive. We store data on it in binary form – ones and zeros. Now, this arm supports a “head”, which is an electromagnet that scans over the disk and either writes data by changing the magnetization of specific sections on the platter or it just reads the data by measuring the magnetic polarization.
Now, in principle, pretty simple, but in practice a lot of hard core engineering. The key focus lies in being sure that the head can precisely, error free, read and write to the disk. The first order of business is to move it with great control. To position the arm engineers use a “voice coil actuator”.
The base of the arm sits between two powerful magnets. They’re so strong they’re actually kind of hard to pull apart. There. The arm moves because of a Lorentz force. Pass a current through a wire that’s in a magnetic field and the wire experiences a force; reverse the current and the force also reverses. As current flows in one direction in the coil the force created by the permanent magnet makes the arm move this way, reverse the current and it moves back.
The force on the arm is directly proportional to the current through the coil which allows the arm’s position to be finely tuned.
Unlike a mechanical system of linkages there is minimal wear and it isn’t sensitive to temperature. At the end of the arm lies the most critical component: The head. At its simplest it’s a piece of ferromagnetic material wrapped with wire. As it passes over the magnetized sections of the platter it measures changes in the direction of the magnetic poles. Recall Faraday’s Law: A change in magnetization produces a voltage in a nearby coil. So, as the head passes a section where the polarity has changed it records a voltage spike.
The spikes – both negative and positive – represent a “one” and where there is no voltage spike corresponds to a “zero. The head gets astonishingly close to the disk surface 100 nanometers in older drives, but today under ten nanometers in the newest ones. As the head gets closer to the disk its magnetic field covers less area allowing for more sectors of information to be packed onto the disk’s surface.
To keep that critical height engineers use an ingenious method. They “float” the head over the disk. You see, as the disk spins it forms a boundary layer of air that gets dragged past the stationary head at 80 miles per hour at the outer edge. The head rides on a “slider” aerodynamically designed to float above the platter. The genius of this air-bearing technology is its self-induced adjustment. If any disturbance causes the slider to rise too high it “floats” back to the where it should be.
Now, because the head is so close to the disk surface any stray particles could damage the disk resulting in data loss. So, engineers place this recirculating filter in the air flow; it removes small particles scraped off the platter.
To keep the head flying at the right height the platter is made incredibly smooth. Typically this platter is so smooth that it has a surface roughness of about one nanometer. To give you and idea of how smooth that is let’s imagine that this section is enlarged until it’s as long as a football field – American or International – the average “bump” on the surface would be about three hundredths of an inch.
The key element of the platter is the magnetic layer, which is cobalt – with perhaps platinum and nickel mixed in. Now this mixture of metals has high coercivity, which means that it will maintain that magnetization – and thus data – until it is exposed to another powerful magnetic field.
One last thing that I find enormously clever. Using a bit of math to squeeze up to forty percent more information on the disk. Consider this sequence of magnetic poles on the disk’s surface – 0-1-0-1-1-1. A scan by the head would reveal these distinct voltage spikes – both positive or negative for the “ones. We would be easily able to distinguish it from, say, this similar sequence. If we compare them they clearly differ.
Engineers, though, always work to get more and more data onto a hard drive. One way to do this is to shrink the magnetic domains, but look what happens to the voltage spikes when we do this. For each sequence the spikes of the ones now overlap and superimpose giving “fuzzy” signals. In fact, the two sequences now look very similar. Using a technique called Partial Response Maximum Likliehood engineers have developed sophisticated codes that can take a murky signal like this, generate the possible sequences that could make it up and then choose the most probable.
As with any successful technology, these hard drives remain unnoticed in our daily lives, unless something goes wrong.