Recollections of BBC engineering from 1922 to 1997
The British Broadcasting Corporation
web site is: www.bbc.co.uk
An Introduction to Broadcast Technology - Fundamentals
The basic requirement of Television sound and Radio is that the listener hears a good reproduction of the original speech, music or other sound generated during a programme. This is usually achieved by using a microphone, transmission system and loudspeaker.
A microphone produces an electronic signal representing the pressure of air surrounding it and since sound is caused by variations in air pressure, the signal represents the sound being picked up by the microphone.
Fig 1 Audio signal from a microphone
It is an analogue signal because its voltage follows the instantaneous air pressure at the microphone and it is often called an audio signal because it represents sound. A loudspeaker does the opposite of a microphone and converts the electronic signal back into variations in air pressure i.e. the original sound.
Transmission is the process of reproducing a signal some distance away from its point of origination. This can be achieved in many different ways as described later in these notes.
Signals can be sent in the above analogue form over short distances without any difficulty, but as the distance gets greater the signal becomes distorted and is no longer a faithful representation. In practice, analogue transmission systems are used a great deal in broadcasting and signal distortions are minimised through careful design.
In order to avoid the effect of signal distortion on analogue transmission systems a different approach can be used. This involves making regular measurements of the analogue signal's voltage, representing each result as a number, sending the number to the destination and converting back to form the original analogue signal. Any distortion in the transmission system is of no consequence provided that the correct numbers are received.
Fig 2 Conversion from an analogue to a digital signal
More details are given in the section on transmission systems, but for now note that a NICAM Coder converts an analogue signal into digital form and a NICAM decoder does the reverse. (NICAM is one of many systems that do this, some of which are described later). Video coders and decoders do likewise for video.
When we hear sounds in real life, the sound arriving at each ear is subtly different. Small differences in echoes and transit time allows the brain to perceive breadth and depth to sounds. When we listen to sounds produced by loudspeakers a minimum of two, spaced, loudspeakers are required to produce a similar breath of sound. This is referred to as stereo sound. Virtually all VHF radio services are broadcast in stereo and require two audio signals of the type outlined above.
Apart from sound, the basic requirement of television is that the viewer sees a good reproduction of the original scene generated during a programme. This is usually achieved by using a camera, transmission system and display.
A television picture is made up of 25 frames (still pictures) a second, which is fast enough to give the illusion of a continuously changing scene. Each frame consists of nearly 625 horizontal lines which are sufficiently narrow and closely spaced to give the picture acceptable definition - i.e. to provide sufficient picture detail. Each line varies in brightness to "paint" the picture.
Fig 3 Four frames of a TV picture illustrating the line structure and a moving image
On a colour television this process happens three times, simultaneously within each frame, to produce separate red, green and blue pictures which are mixed together to produce the required range of colours. Close examination of a television screen will reveal a repeating pattern of small red, green and blue stripes or dots which, when viewed at a distance, appear to merge together.
Illustration to show how a colour picture is formed
A television camera works by scanning the scene from side to side and top to bottom, covering a complete picture 25 times a second. At any instant, the camera is only looking at one small spot on the scene and the intensity of light at this particular point is converted into a voltage. Therefore as the camera scans the entire scene, its voltage output varies in sympathy with the brightness of the light at the point concerned.
A television display tube does the opposite of a camera in that it scans the picture from left to right and then top to bottom, producing light of varying intensity to build up the picture. As the spot moves across the screen a voltage, equivalent to the voltage from the TV camera, is applied to the television display tube to vary the brightness of the light.
When the camera starts to scan the scene (top left hand corner) it must signal this fact to the television set so that the scan of its display tube also starts on the top left hand corner in synchronism. In order to achieve this, the camera produces a special signal called a synchronising pulse, or sync pulse, which is detected by the television set. In fact there are two types of sync pulse, one to identify the scan starting at the top of the picture and the other to identify the start of each line across the picture. These are called field syncs and line syncs respectively.
In order to understand the process, it is easier to consider line syncs to begin with. As the scan progresses across the picture, the camera produces a voltage which varies according to the intensity of the light and typically this would be between 0 volts for black and 0.7 volts for white, with shades of grey being in between. On reaching the right hand side of the picture, it takes a finite time for the next scan to start on the left hand side of the picture. During this time, the camera produces a different voltage which can be distinguished from the rest of the signal and in practice this is usually minus 0.3 volts. The result is a signal which varies, typically, as follows:-
In the television receiver, each sync pulse is detected and used to trigger the start of each scan across the picture.
Although some modern display systems scan the picture progressively from top to bottom as outlined above, analogue television systems transmit a signal that facilitates interlaced scanning, in which a frame is made up of two fields of lines that are offset as illustrated below. Interlaced scanning was introduced to reduce ‘flicker’ which would otherwise spoil the viewed image as the 25 pictures per second are built up.
Fig 6 Illustration of interlace scanning
Field syncs are similar to line syncs (-0.3 volts) but of longer duration and these are used to start the scan of each field at the top of the picture. Reference was made earlier to the picture being made up of "nearly" 625 lines The reason is that although there are 625 lines in a frame, not all of them are displayed because, in the early days of television, a significant time was required for the spot to travel from the bottom right hand edge of the screen to the top left. This time is now used to carry data signals (see Teletext).
Fig 7 Field syncs and Teletext
The numbers identify individual lines at the beginning of the first field, starting with the lines containing field syncs. Lines 8 to 16 contain teletext data, there is an Insertion Test Signal (ITS) on line 21 and the first line of the picture is 23. The ITS is normally generated at a studio and measured at various points in the programme chain, including the output of a transmitter, in order to measure the amount of distortion that has been introduced.
Colour pictures start life in the camera as three separate "components" for red, green and blue. Each of these components is in the form of a signal similar to that illustrated in Fig 5 (but usually without the sync pulses). These three signals are then processed in a fairly complex way to produce a single signal which combines both brightness and colour information (more correctly described as luminance and chrominance). In the UK, the signals are encoded using the PAL system which stands for Phase Alternate Line.
Compared with Fig 5, Fig 8 shows the effect of adding colour. Broadly speaking, the thicker the line the deeper the colour, or put more correctly, the higher its saturation. When the magnified part of the signal moves sideways the hue of the colour changes e.g. orange to yellow. This signal is called the colour sub-carrier. The colour burst provides a reference for the colour sub-carrier which enables TV receivers to reproduce the correct colours (it's a sort of sync pulse for the colour part of the signal).
Video signal transmission is similar in many respects to audio signal transmission The main difference is that a video signal needs more bandwidth, which is roughly analogous to a motorway compared with a country lane. The concept of bandwidth is covered later in these notes.
This is a measure of the rate at which the instantaneous voltage of a signal varies throughout a complete cycle. A signal representing a pure audio tone might vary from 0 volts, up to +2 volts, down through 0 to -2 volts, then back up to 0 volts, 500 times in a second.
Such a signal would have a frequency of 500 Hertz (1 Hertz = 1 Hz = 1 Cycle per second).
signals range from about
30 Hz to
signals range from about
0 Hz to
signals range from about
15 kHz to 100,000,000,000 Hz
The Broadcast radio signals range from 198 kHz to 14 GHz
(Radio 4 from Droitwich) (satellite systems)
The list above illustrates several abbreviations that are commonly used in transmission engineering and here are some more:-
Each prefix may be used to qualify various different units such as:
= cycle per second - a
measure of frequency
So, for example, 1kW means 1000 Watts and, as indicated above, 20kHz means 20,000Hz..
The size of a signal and its frequency can be shown in graphical form (like a graphic equaliser display on some Hi-Fi systems):-
Fig 10 A 500 Hz pure tone (as shown in Fig 9) displayed in frequency spectrum form
This type of diagram shows the spectrum of a signal - a very important concept in transmission engineering. The 'size' of a signal is represented by its power, which is related to its voltage. The pure 500Hz tone is shown as a single vertical line but an audio signal would occupy a range of frequencies:-
Fig 11 The spectrum of an audio signal
In practice the power of an audio signal at different frequencies varies with time (e.g. during a passage of music). However this diagram shows the power that could be present.
Note that although the audio on Compact Discs has a spectrum up to 20kHz, broadcast transmission systems are usually limited to 15kHz or less.
Spectrum diagrams appear several times in these notes, but with various different scales.
Radio waves or, more correctly, electromagnetic waves, are produced when a suitable signal is applied to an antenna (often called an aerial):-
Fig 12 Transmission of electromagnetic waves from an antenna
As indicated above, this only works when the signal is between about 15kHz and 100GHz. It is also necessary for the length of the antenna (between a and b) to be around half the wavelength of the signal.
So, a radio broadcast on 100MHz would have a wavelength of 3 Metres and ideally the length of the antenna would be 1.5 Metres. More precisely, the length of a "dipole antenna", as shown above, would need to be about 1.5 Metres, but an antenna could have a number of such dipoles (or similar radiating elements).
If an audio signal with a spectrum of say 30Hz to 10kHz is applied directly to an antenna, radio waves are not produced because the frequency range is too low. Therefore a different approach is needed.
This is a process in which one signal is modified by another.
The signal to be modified is called a carrier and it has a frequency suitable for feeding an antenna. This diagram shows a carrier.
The signal that we wish to convey may be a simple audio signal as shown below (pure tone or "sinewave"):-
In order to convey this audio signal it modifies the carrier by changing its amplitude as shown below:-
Alternatively the audio signal can be made to modify the carrier's frequency:-
Fig 13 A carrier being a) amplitude modulated and b) frequency modulated
Modulation has an effect on the spectrum of a carrier. Consider first an unmodulated carrier:-
Fig 14 An unmodulated radio frequency carrier e.g. Radio 4 during a silent period!
When it is modulated, power is produced from the transmitter not only at the carrier frequency, but also over a band of frequencies above and below the carrier frequency. In the case of AM the power extends away from the carrier frequency in "sidebands" by an amount equal to the highest modulation frequency - say 5kHz in this case.
Fig 15 An amplitude modulated radio frequency carrier e.g. Radio 4 without silence
In the case of FM the sidebands extend considerably further.
Because all transmissions have sidebands, the carriers have to be spaced far enough apart to avoid interference.
Fix 16 Interference caused by sidebands overlapping
Since there is a finite amount of spectrum available, this limits the number of transmissions. However transmitters can use the same frequency for different programmes provided that they are sufficiently far apart, so that the interfering signal is very weak.
The allocation of frequencies to transmitters is called service planning. It is a complex process, but one simplification is that all transmissions fit into channels, which are standardised ranges of frequencies suitable for the output from one transmitter. Confusingly BBC1, BBC2, ITV and C4 are often called channels. However taking just BBC1, this is transmitted on a number of radio frequency channels in the UHF band. e.g. the Sutton Coldfield transmitter broadcasts BBC1 on channel 46 and Crystal Palace transmits BBC1 on channel 26.
All signals which are received by a radio or television set are, in effect, mixed together so the spectrum at the input of a radio might look like this.
Fig 17 Several channels being received on a radio's antenna
A radio is sensitive to one channel at a time and the required channel can be selected by turning the tuning knob. The dotted line above shows one possible tuning position.
On the face of it, the radio frequency or RF spectrum seems very large, but in practice the usable spectrum is quite limited by propagation characteristics. This refers to the way in which the signal travels from the transmitting antenna and the situation varies with the radio frequency being used. The frequency spectrum is divided into a number of 'bands', each of which tend to have different propagation characteristics, as summarised below. Each band has a name which refers to its range of frequencies and some have another name which refers to the range of wavelengths although this is no longer the preferred terminology:-
30 kHz to 300 kHz signals propagate over long distances (1,000 kilometres) and follow the curvature of the earth. e.g. the BBC Radio's 4 transmitter at Droitwich in the Midlands which can be received in Ireland.
1.9.2 Medium Frequency (MF) or Medium Wave (MW)
300 kHz to 3 MHz signals don't travel so far but they do follow the curvature of the earth. Unfortunately, especially at night, they can travel further by going towards the sky and being reflected back down again. This happens in a rather uncontrolled manner and causes interference.
3 MHz to 30 MHz signals only travel a short distance along the ground but they can travel thousand of miles by going up to the sky and being reflected back down again from the ionosphere, which is a few hundred km above the earth's surface.
30 MHz to 300 MHz signals are too high in frequency to be reflected back from the ionosphere but they can be received up to about 80 km from a large transmitter (the lack of sky wave interference helps in this respect). e.g. Radio 2 on VHF/FM from Wrotham in Kent which can be received in Hertfordshire.
300 MHz to 3,000 MHz signals are similar to VHF except that the signals tend to get blocked by hills and large buildings. This is why there are very many UHF TV relay stations.
3 GHz to 30 GHz signals only travel over a line of sight. e.g. satellite television and point to point telecommunication systems that use microwave dishes to send signals between one tower and another up to about 70 kilometres apart.
In general, large areas can be covered by high power transmitters and small pockets of population which are screened by hills etc. can be covered by a much larger number of low power transmitters. Radio frequency power feeding a transmitting antenna is measured in watts (W) or kilowatts (1kW=1000W), just as it is for electric lamps and fires. A small TV relay station serving a population of 300 people would typically have a power of 2 watts, whereas a high powered station serving millions of people would typically have a power of 20 kW. HF transmitters have powers up to 500kW.
Amplifiers are needed in all transmitters as well as in many other broadcasting systems. Fundamentally they take a small signal and make it bigger, so for a small range of input voltages an amplifier should produce a larger range of output voltages as illustrated below. The graph headed 'linear amplifier' shows the output voltage on the vertical axis, for a given input voltage (horizontal axis). The steeper the slope, the greater the amplification (for given scales on the graph).
Fig 18 A signal passing through a perfect linear amplifier
The spectrum of the output signal is the same as the input, but larger.
In reality it is very difficult to make a perfectly linear amplifier, especially at the high powers and high frequencies involved in broadcasting. When an amplifier is not perfectly linear, as indicated by the curved line below, then the output becomes distorted.
Fig 19 A signal passing through a non-linear amplifier
A consequence of the distortion is that the amplifier generates new and unwanted frequencies at its output. Furthermore, if there is more than one input frequency, some of the unwanted frequencies will be very close.
Fig 20 Generation of Intermodulation Products
These unwanted frequencies are called intermodulation products or IPs and must have a power typically no more than about a millionth of the power of the wanted signal.
(The above explanation refers to voltage amplification in order to avoid references to current and impedance. However most amplifiers increase the power of a signal. Note that power is produced when a voltage appears across a resistive load such as an antenna, or an electric kettle!).
Terms like 'a millionth of the power' are rather unwieldy so powers are usually compared using decibels, or dB for short. This unit of measurement is defined using a mathematical formula, but engineers usually remember a few basic results of this formula:-
twice the power
So, an amplifier that increases the power by 10000 has amplification (usually called 'gain') of 40dB. An advantage of this system is that when the output of one amplifier is connected to the input of another the overall gain is the sum of the individual gains. e.g. an amplifier that increases the power ten times, followed by an amplifier that doubles the power, would give an overall gain of 13dB.
The decibel is a relative measurement, but the term is often used to quote an absolute quantity and probably the most familiar refers to sound levels. This involves relating a measurement to a known quantity which, in this case, is the 'threshold of hearing' (the quietest sound that can be heard). In the case of transmitter systems the known reference quantity is often 1mw of power, so 1kw would be 60dBm, where the 'm' means relative to a milliwatt.
Returning to the end of the last
section, engineers would say
Multiplexing arises in many aspects of transmission and fundamentally it is a technique for mixing signals together in such a way that they can easily be separated. There are two basic types, frequency division multiplexing and time division multiplexing.
1.13.1 Frequency Division Multiplexing
This has been covered already! When a transmitting station broadcasts on several different frequencies simultaneously, they are all picked up together on the antenna of an radio and then the required signal is demultiplexed by tuning to the appropriate frequency.
1.13.2 Time Division Multiplex
This normally applies to digital signals, which are explained in more detail later, but in essence it involves interleaving several signals to form one composite signal and then reversing the process at the receiving end. Each input signal is selected in turn as illustrated below.
Fig 21 Illustration of Time Division Multiplexing