answersLogoWhite

0

Magnetic storage devices are based upon the principal of electromagnetism. As an electric current passes through a conductor, a magnetic field is created around that conductor. The magnetic field rotates clockwise or anti-clockwise around the conductor depending on which direction the electrons flow through the conductor.

When a conductor is connected to a DC battery, the electrons flow from the negative terminal to the positive terminal which is the complete opposite of how we imagine they travel. However, this imagined direction gives us a simple but literal "rule of thumb" to determine the actual direction of the magnetic field. If we were to grasp the conductor in our right hand such that our thumb pointed in the imagined direction the electrons were flowing (towards the negative terminal), then our fingers wrapped around the conductor would tell use the direction the magnetic field was travelling. If we were to use the true direction of electrons (negative to positive), then we would use our left hand with our thumb pointing towards the positive terminal.

In order to switch the rotation of the magnetic field from clockwise to anti-clockwise, we need to switch the direction the electrons travel through the conductor. Put simply, if we swap the ends of the conductor, the electrons will travel in the opposing direction and thus switch the rotation of magnetic field.

Now, the really interesting part about electromagnetism is not that an electric current can create a magnetic field around a conductor (although that is quite interesting in and of itself). No, the really interesting part is that when we bring a conductor into the proximity of a magnetic field (or indeed bring a magnetic field into the proximity of a conductor) we create an electric charge within the conductor! The polarity of the electric charge (that is, the direction in which the electrons travel through the conductor) depends upon the motion of the magnetic field relative to the conductor. As they move closer, we get one polarity, while pulling away we get the other.

Now , take a moment to let all this sink in, it's a vital life skill! Not only can we use electricity to create a magnetic field but we can use a magnetic field to create electricity! More importantly, the polarity of the electricity dictates the polarity of the magnetic field, and vice versa. Electromagnetism is not just a big fancy Greek word: electricity and magnetism are naturally symbiotic!

In order to use electromagnetism as a means of storing data, we first need a medium that can physically record these magnetic fields. Magnetic tape is an obvious example. This is really nothing more than a non-magnetic "substrate", typically Mylar, coated in a layer of magnetic material primarily composed of iron oxide. As the tape passes through the magnetic field generated from electricity, the iron oxide particles re-align themselves according to the relative strength and polarity of the field and will maintain that state until we apply another magnetic field to it. Since the tape is moving through the magnetic field at a constant speed, it will faithfully record every change that occurs to the magnetic field.

Once we've completed a recording, we will naturally want to replay it. To do so we simply reverse the process. As the tape passes over the conductor, the magnetic fields upon the tape reproduce the electric current that was used to create the magnetic fields in the first place. As the magnetic field changes in strength and polarity, so the electric current changes to suit. Thus we can faithfully reproduce every nuance of the original electric current. Well, not exactly, there's always a little lost along the way, but its close enough as makes no practical difference.

an obvious practical application is that we can use electromagnetism to make an analog recording, such as when recording a vocal performance. A microphone converts the sound waves produced by the performer into an oscillating electric current that is analogous to the original sound wave. That electric current is then used to generate oscillating magnetic fields which can then be recorded onto magnetic tape. When the recording is complete, we can replay the tape to recreate the electric currents which can then be passed to a speaker which converts the electrical energy back into sound energy.

This is an over-simplification. However, if we can use electricity and magnetism to record and reproduce something as complex as an analog sound wave with a reasonable degree of accuracy, then recording digital information should be a walk in the park.

Unfortunately it is not, but it's also not overly difficult. Whereas analog recordings try to faithfully reproduce all the subtle nuances of a sound wave, digital recording are only interested in reproducing sequences of 1s and 0s. Electromagnetism involves negatively and positively polarised electrons, so it seems obvious that we should use negative polarity to represent 0s and positive polarity to represent 1s. If only it were that simple...

The problem with this method is that we cannot record long sequences of consecutive 0s or 1s because there's simply no way to differentiate one bit from the next in the absence of a magnetic flux reversal. That is; a transition from 0 to 1 or vice versa. If we write 100 zeroes, we want to ensure that when we read it back we get exactly 100 zeroes, not 101, not 99, but exactly 100.

To get around this we use the transitions themselves to denote the 1s and 0s, never the polarity. In this way we are no longer concerned with the polarity, only in whether or not a flux transition has occurred (regardless of which direction). Now, if we say that a flux transition now represents a 1, then we can guarantee that a long sequence of 1s will be represented by an equally long sequence of alternating flux transitions. That's half the problem solved at a stroke. That just leaves the zeroes which we must denote with a non-transition. To resolve that problem we simply insert periodic clock transitions. That is, if we insert a clock transition in front of every bit, then a 1 becomes two consecutive transitions while a 0 becomes a transition followed by a non-transition. Problem solved!

Actually, no, it isn't. The problem with this encoding is that the minimum run of non-transitions is 0 while the maximum is 1. Technically this is a variant of Run Length Limited encoding known as RLL 0,1, but is more commonly known as Frequency Modulation (FM) encoding. We haven't used FM encoding since the 70s and for good reason. There's inevitably going to be a lot of transitions, at least one in every data bit boundary, and every transition slows down reading and writing and reduces overall capacity.

If we can reduce the number of clock transitions, we can not only increase read and write speed, we can also increase capacity by packing more data bits into the same space. We can easily achieve that by saying that a clock transition should only be inserted when a 0 immediately follows a 0. If we then say that a 1 is always a non-transition followed by a transition while a 0 following a 1 is two non-transitions, then a 0 following a 0 becomes a clock transition followed by a non-transition. In this way the minimum run of non-transitions is 1 and the maximum is 3 (technically known as RLL 1,3). In so doing we've effectively halved the number of transitions, doubled the clock frequency and thus doubled the capacity. Virtually every floppy disk and hard-drive up until fairly recently used RLL 1,3, more commonly known as Modified Frequency Modulation (MFM) encoding.

Today we predominantly use RLL 2,7 encoding although this has proved unreliable with today's highest capacity drives which now predominantly use RLL 1,7 encoding. Both offer improved data density and speed over MFM.

User Avatar

Wiki User

10y ago

What else can I help you with?