In this, the second part of this set of articles, we will look at recording and playback heads and how they are designed, and how the design/construction affects the operation.
To review: If we bend a bar magnet into a U shape we have the typical horseshoe magnet. Since we can also make something magnetic but applying a current to a wire wrapped around a magnetizable materials. If we wrap wire around a bar it will become an electromagnet when we apply power. [Electromagnets are only magnetised during the time we apply the current] If we bend this bar into a U shape we have an electromagnetic equivalent of the typical horseshoe magnet.
A recording head is a horseshoe magnet taken to the extreme. The difference is that the open end of a recording head is extremely small. The easiest way to understand a very basic recording head is take a ring of metal, and cut a very small slot through it. Then we wrap some turns of wire around this ring on the side opposite the slot. Connect the wire it to an electron source, we will have a magnet with the magnetic field focused at that very small gap opposite the coil of wire.
That's basically all that a recording head is. [This tutorial on recording/playback heads - showing their construction - will focus entirely on this type of head called an 'induction' head. We limit this tutorial for purposes of understanding how these work, though newer type of heads exist and this style of head isn't used much anymore. These heads are just like the heads in your audio cassette recorder.]
The more turns of wire we make around the end opposite the gap the the larger the field we can generate at the gap. We are also limited by how far apart the ends are at the slot we cut. We call this slot the 'head-gap'.
The size of this gap determines how large an area of the tape we will magnetize, and it also determines the smallest area of magnetization we can resolve when used for reproduction. The gap size used in the design of the tape head needs to match the type of magnetic media we will use. We'll discuss those requirements in the next article.
Imagine the magnetic field at the gap as lines of force between the ends of the gap. But on the edge of the gap - the part where the tape will touch - these fields start expanding outward like waves in a pond. [ I really need to design a graphic to show this so I hope you can visualize this from my description].
Most people have noticed that if you drop a rock in water there will be waves going out in circles from the point the rock hits. If you dropped the rock in a swimming pool at the edge of the pool then these waves would be half-circle spreading out from the edge.
Keeping this model in mind, then you should be able to visualize a magnetic tape with these waves going out from the gap and into any material which would be touching this gap. That's how we magnetize the tape so we can store magnetic images of our electric energy on this tape.
Remember that we generate an electric current by moving either the wire in the magnetic field or by moving the electric field in respect to the wire.
If we had a tape moving across the head while we changed the electric current in the coil of wire wrapped around the head, changes in the magnetic field would generate corresponding changing magnetic images would on the tape.
Once we have done this, we can rewind the tape, and disconnect the head from the current generator. Next we then connect these wires to a voltmeter or amplifier, and then move the tape across the head in the same way we did during the recording.
As the tape moves across the gap the magnetic field will be induced into the head [that's where we got the term induction head above] by the changing magnetic field and will cause a current flow through the wire coil on the far side of the gap.
This will be a relatively small current when compared to the current which was used to generate the field which magnetized the tape.
The level of this current generated by the head and the sensitivity of the amplifiers receiving this signal are just two factors which determine how effectively we can use magnetic media to store electrical changes.
Other factors that come in to play are how fast the tape moves in relation to the head gap. This is a key point - how fast the tape moves in relationship to the GAP. Remember this. Other things which need to be considered are the amount of wire on the coil, the area over which the tape passes, and the gap width of the head. [By width I mean the distance between the end of the pole-pieces made when we cut the slot]
More wire will cause a larger current to be generated. The area of the tape and the area of the head it is passing over also cause more current generation. The wider the tape the more magnetic field it will generate. This also is dependent on how wide the gap is. The smaller the gap the smaller the current.
In recording side of this discussion the smaller the gap the smaller the magnetic field which will be generated. So to get a larger signal you use a wider gap - within limits.
If you make the gap wide the data signal takes up more room on the tape and the less data you can store, as fewer pulses will fit on the tape.
This brings up yet another design factor. To generate a signal in the head we have to generate a magnetic field. The speed at which we move the tape is also dependent on how large the gap is and how fast we are sending data signals - eg: how fast our magnetic field is changing. Everything interacts - that's what makes it hard to understand easily.
We are going to use some simple math here. The figures do not relate to any real tape device, but are chosen to make is very easy to illustrate and understand. But first one more bit of vital information is needed.
For purposes of this explanation - though not entirely accurate for modern magnetic devices we shall assume that if we have 1000 bits of information we need to have 1000 pulses [separate magnetic fields] on the tape. This is because we can not record a continuous non-changing level on the tape otherwise the magnetic level would not be changing with respect to the head.
If send a steady [non-changing] signal to the head, we will orient all the particles on the tape in the same direction. When we played the tape back, as the first part of the signal was induced into the head we would see a rise in output voltage, and then a decrease to zero, as we are no long changing the field. Remember we have to have movement of a wire or field to generate the current. If all the fields are oriented in the same direction we will see this movement only for a short time then the field will then be at a steady state.
Lets assume we send these 1000 pulses in one second. That means each pulse is 1/1000th of a second long in the time domain.
To be recorded properly a changing magnetic field must span the gap. If the field were the same size as the gap the level would be too low and if the field were shorter than the gap then it would be 'invisible' to the had as the magnetic field would be shorter than the gap. You could say it got lost in the gap.
The gap needs to be 1/2 the size of the highest frequency you wish to reproduce. Any larger than this and the output starts falling off sharply until they are the same size when output is virtually non-existent.
If we move the tape at 1 inch per second then the gap needs to be 1/2 of the pulse length or 1/2000th of an inch. If the gap were 1/1000th of an inch then we'd have to move the tape at 2 inches per second. The latter will give higher output at the expense of cutting the capacity of the tape by one half.
We will discuss the way the design of the magnetic media itself affects this in the next part, but as far as heads go, you can see we have several trade-offs, all of which are interdependent.
If we make the record head gap smaller we can record more information, but we have cut the amount of energy we can impart onto the tape. We also reproduce the signal with this smaller head gap so that means we cut the output even further. If we slow down the tape we get even lower output. [Data recording typically uses the same head for recording and playback, as opposed to audio recording, where tolerances are more critical for their purposes].
In a later segment, after the media upon which we record is explained, you will see how data capacity was expanded by several different methods, than just speed and gap size.
Got something to add? Send me email.
More Articles by Bill Vermillion © 2009-11-06 Bill Vermillion
Every piece of software written today is likely going to infringe on someone else's patent. (Miguel de Icaza)