High Quality Sound production and Reproduction

High Quality Sound production and Reproduction
About this book
T w book was produced in the Central
Programme Operations Department of the
BBC for their own personnel, both technical and non-technical, to enable them
to obtain the best results from studio
equipment. But it will also prove of
absorbing interest to other broadcasting
organisations and people, both amateur
and professional, interested in the production and reproduction of highquality
The book is divided into three parts,
the first of which deals with the basic
principles of sound and electricity and
includes chapters on the theory of musical
instruments and studio acoustics.
The second part describes studio equipment. Microphones, loudspeakers, studio
control desks, outside broadcasting equipment and P.A. equipment arc among the
topics fully discussed.
The finalsection deals with the important
subject of the placing of microphones for
talk, vocalists and all types of bands,
orchestras and musical groups. Other
subjects covered are the control of volume,
the production of sound effects and sterwphony.
The book is lavishly illustrated throughout with half-tones and line drawingsthe latter bringing a novel approach to
the pictorial presentation of complicated
information in a most simple form.
274 pages, inclrtiling 175 dragranls in thr
text, plus 46 p a p s qi'art plates.
sound technical
Britain's foremost journal for
electronics, radio and television
WORLD.I t provides
a n authoritative survey of
international progress.
Leading specialists describe
developments in both theory and
practice. Problems of designing
circuits for current applications
are discussed in detail and
illustrated with typical examples.
WORLDalso covers
important conferences and
exhibitions-in Britain and
abroad-reviews latest equipment,
and reports news of the electronics
and radio industry.
The first journal ever devoted
to radio, founded more than fifty
years ago, W m ~ t ~ WORLD
today a n essential source of
technical information for all in
electronics and telecommunications.
Monthly 2s 6d Annual subscription f 2
W ireless World
from all ncwsngen~sor direct from
42s net
The first three chapters are given up to a brief summary of
those aspects of acoustic theory which are most useful in the
business of collecting sounds in the studio and reproducing them in
the listener's home. We begin by building up a picture of sound
Fig. I . I . Hmtmt~ntol
m o m of a wolcr
wane rcrulting from
ih mtical mbratwn
of individual pmtich
waves in action and defining some of the technical terms in common
use. Consider first the mechanism of waves on the surface of water.
Drop a stone into a still pond (the analogy in terms of sound would
be a hand-clap or an explosion) and a wave will be seen to travel outwards in an ever-increasing circle, the wave taking the form of a
disturbance on the water's surface. Drop a succession of stones, or
plunge something up and down in the water (analogous with a
continuous sound), and a continuous rippling outwards will take
place, capable of setting a line of small corks into upand-down
In Fig. 1.1, the wave is represented in successive drawings as
moving to the right, while the corks-which we have introduced
simply as-markers or guides to the water vibration-move up and
down on vertical lines just as if they were suspended from a spring.
The natural time delay for greater distances from the source means
that successive corks reach the top of their swing a little later, so
that the crests of the wave appear to travel outwards. In fact, there
is no outward movement of the corks and water at all.
Set a tuning-fork in vibration and waves of energy will travel
outwards. This is a sound wave. I t takes the form of a disturbance
in the air (or other medium through which it travels), and is
capable of setting a thin membrane, such as the ear-drum, in
to-and-jio vibration. As with the water waves, the air particles do
not travel out with the wave. They imitate the fork vibrations,
oscillating " on the spot ",and pass on the energy of the wave by
"On Illc spot" vibralnr of air
fiarticlcs in a sound waue
reason of the elastic coupling that exists between them (shown in
Fig. I .2 as coupling springs).
The corks which we used as guides to the water vibration are not
going to be of much use to us here, but the history of positions taken
up by the tuning-fork may be recorded by fastening a tiny pen to it,
and moving paper underneath. The wavy trace, by introducing
the appropriate delay in time, might be a record of the movements
of any of the air particles in the path of the wave.
Fig. 1.3 will repay closu examination. Starting at the middle
point, A, we see that the fork moved out to the right, then swung
Fig. 1.3. Wmcfmm& o d d by kming-f~k on moving paw
back through the midpoint and fully out to the left. It then returned to the midpoint, at B, and began to repeat this sequence of
movements all over again. (The result is a sine wave, showing that
the fork is performing simple harmonic motion, as described in the
The Frequency of the fork vibrations is defined as the number of
complete vibrations (or cycles) per second. Thus, if NO complete
vibrations are performed in one second, the frequency would be
written 40CIS. Where very high frequencies are being considered
we use kilocycles per second and megacycles per second as units.
Thus, 5,000,000 c/s may be written as 5,000 kc/s or 5 Mcls.
In point of fact, although vibrations at any frequency will give rise
to sound waves, the human ear is only sensitive to sounds in the
frequency range 20 c/s to 20,000 c/s approximately. More will be
said about this in a later chapter.
The pitch of a musical note may be defined as that property by
which we place the note as being high or low on the musical scale.
Pitch and frequency are related, and increasing or decreasing the
rate of vibrations will cause the sound to move up or down the
musical scale. For example, by agreement in 1939, Standard
Musical Pitch A, which is shown in musical notation in Fig. 1.4, has
been fixed at 40c/s, and any structure vibrating regularly at this
rate, violin string, air column, or circular saw, will emit this note.
A discussion of the connection between frequency ratios and
musical intervals will be reserved for the next chapter, but the simple
rule of doubling the frequency to raise the pitch by an octave will
already be known to most readers-eg., the A above Standard
Musical Pitch A is 880 c/s.
The tuning-fork, an invention of John Shore in I 7I I , is well known
for its ability to maintain vibrations purely at one frequency. Most
other sound sources, musical instruments included, tend to perform
several modes of vibration simultaneously, so that the sounds which
result are not pure tones, but complex. Usually the various component frequencies form a family, or series, being simple multiples
(twice, thrice, etc.) of the lowest frequency present. This last is
Fig. 1.5. Firsr tight harmonics
of rro c/s shown in musical
called the fundamental jkquntty, and it decides the apparent p h h
of the note. The higher frequencies, or overtones, are called
Thus, a double bass sounding A two octaves below 440 c/s will
contain, mixed in certain proportions, the following family of
2nd harmonic
3rd harmonic
4th harmonic
5th harmonic
I 10
330 4 s
440 c/s
550 c/s
and so on (Fig. I -5).
It is the presence of harmonics in varying numbers and relative
strengths which give the sounds from different musical instruments
their characteristic quality or timbre. We recognise the clarinet,
for example, partly because of the preponderance of odd harmonics
(third, fifth, etc.) which that instrument produces.
The waveform corresponding to the pure vibration of a tuning-fork
was seen in Fig. 1.3 to be a simple sine wave. When several frequencies are present together the resultant waveform, if displayed on
Fig. 1.6. Wavcfmproduced by combiningfundamental
and 2nd harmonic
a cathode-ray tube, for example, is found to be more complex.
Generally speaking, the peakier the waveform, the higher the
harmonic content. A saw-toothed waveform may contain the
complete harmonic series.
Fig. 1.6 shows in diagrammatic form the derivation of the waveform produced by a fundamental and second harmonic which are in
phase (for definition see section 1.5) and whose amplitudes are in
the ratio of 3 : I.
When a number of frequencies are sounding simultaneously, the
brain receives the impression that additional frequencies are
present. This effect is inherent in the mechanical action of the ear,
and makes analysis of complex sounds in terms of the sounds actually
imagined in the brain exceedingly difficult. One of the " extra "
sounds created in this way, when two pure tones are being sounded
together, is at a frequency which is the difference between the two
real frequencies. For example, if tones at 1,000 c/s and goo c/s
are present, the hearer feels that I oo c/s ( I ,000 -900) is also sounding.
Thin particular combination tone is called the dz&ence
When two real frequencies differ by only a few c/s, the listener
receives the impression of pulsations at the difference frequency,
which is usually called in this case the beat frequency. The process
of tuning two notes to exact unison is thus one of eliminating beats.
Phase is the term used to describe the stage reached by a vibrating
particle in its cycle of movement. Phase is usually measured in
degrees, 360" corresponding to one complete cycle (see Appendix).
The distance, measured along the wave, between successive
particles which are in # h e , is called the wavelength (symbol A)--e.g.
the distance from crest to crest. Now the wave will travel this
distance in the time it takes the source to perform one cycle, and it
will travel f times this distance in one second, if there are f cycles
Fig. 1.7. Derivation of fornula c =fA
per second. Therefore, if the frequency of the tuning-fork in
Fig. I .7 is f c/s, and the wavelength is X feet, the distance travelled
by sound per second isf x X feet per second.
The distance travelled per second is called the velocity (symbol c ) ,
and so we have the important formula
which is true for all types of waves (Fig. 1.7). The velocity of
sound in air is approximately 1,120 ftlsec, and the formula becomes
r ,I 20 cfh, from which it is possible to calculate the wavelength in
air for any frequency, and vice versa.
Example 4
What is the range of wavelengths in air corresponding to the
range of normal hearing?
Lowest frequency is 2 0 cls
Highest frequency is 20,000 c/s
= 0.67 in.
The veIocity of sound waves is virtually independent of frequency.
If this were not so, the various notes from an orchestra would reach
the audience in a hopeless muddle. It is found, however, that
temperature has a direct bearing on sound velocity, and c increases
by about 2 A/sec for each degree Centigrade rise in temperature.
The velocity at o0 Centigrade is 1,087 ftlsec, and the velocity at any
temperature T can therefore be calculated from the formula
An example of this effect occurs in the tuning of wind instruments. The wavelength of the notes from a wind instrument is
decided by the length of the air column. If the velocity increases
due to warming of the air, this will tend to sharpen the pitch of the
notes. I t is important, therefore, when tuning, that such instruments be brought up to the temperature in which they are to be
played. A temperature change of 15"Centigrade results in a shift
of about one semitone.
The humidity, or moisture content of the air, has only a small
effect on sound waves. The velocity is about 3 ftlsec faster in
saturated air than in dry air.
So far we have restricted our discussion to the rate of vibration, or
frequency, of the source, and the nature and speed of the waves. I t
is now necessary to examine the behaviour of the source when the
strength of the vibrations is varied. One obvious measure of
the strength of the vibration is the distance which particles of the
source swing to either side of their average position. This is called
the amfilitude, and more violent vibrations of greater amplitude
result from striking, bowing, etc., harder, which corresponds to
putting in more energy. This naturally causes a greater amount of
sound energy to be radiated. The amount of energy radiated per
second is called the power of the source.
The intensity of a sound due to a given source is proportional to the
power of the source and inversely proportional to the area over
which the sound is spread.
The loudness of a sound i s dependent on the power of the source,
and on the nature of the particular sound. By " nature" is
meant the total composition of the sound at the position in which
*Loudness as defined above is a quantity which can be measured. The subjective aural loudness actually heard will include these quantities, but will also
drpend on a number of other factors. This "apparent loudness" will change
with the level cf background noise; the particular person hearing it; and the
psychological and physiological condition of the person at the time. These factors,
outside the control of a broadcasting organisation, make the choice of transmission
levels, for example between speech and music, very difficult (see chapter 14).
it is heard; namely, its frequency content, and the effect of the
acoustic environment.
When a sound source such as a tuning-fork or violin-string is
struck, and left to itself, it performs what are called free vibrations.
These will take place at a frequency-the naturalfrequency-which
is determined by the mass and springiness of the source and by
nothing else. In the same way, a pendulum will swing to and
fro at a rate (frequency) which is a function of the length and of
nothing else.
The rate at which the vibrations die out depends on the rate
of expending energy (defined above as the power), which is made
up of the radiated energy plus the energy lost in overcoming air and
other friction--called the damping.
Thus, a tuning-fork held in the hand decays very slowly, and is
said to be lightly damped. When placed on a table or box, the
fork radiates more energy per second, and comes to rest more
rapidly. Similarly, a garden swing will soon come to rest if we
scrape our feet on the ground, and the dampers on a piano silence
the notes almost immediately.
Forced Vibrations and Resonance
If an alternating force of, say, 500 cis is applied to a tuning-fork
whose natural frequency is 400 cis, the vibrations which result
will be at 500 c/s and will be of very small amplitude. These are
called forced vibrations.
If the driving frequency is then gradually lowered, the fork
vibrations will follow it, since all the energy is being supplied by the
driver, and the amplitude of the fork vibrations will be found to
increase. When the driving frequency coincides with the fork's
natural frequency (400 CIS), very great amplitudes will result.
Further reducing the driving frequency causes the fork amplitude
to fall again to a very small value.
The increased response of a vibrating system, whether it be
mechanical, acoustical or electrical, system when driven at its own
natural frequency is called resome, and so the natural frequency
of a system is sometimes called its resonant frequency.
Some examples of resonance are :shattering a wine glass by singing or playing its '' ringing "
(2) starting heavy bells by pulling lightly on the rope once in
each cycle;
(3) unwanted resonance in the air contained in studios at frequencies which are determined by the dimensions;
(4) the boominess of some loudspeaker cabinets.
Shqmess of Resonance
If we plot a graph of amplitude against frequency, a peak is
found at resonance with falling response on either side. The
sharpness of this resonant peak depends on the amount of damping
A lightly damped system is said to be
Fw. 1.8.
Simp and bmad rdJanrmrc
shurpb tuned, and will respond only to frequencies near resonance
e.g. a tuning-fork.
A heavily damped system is broadly tumd, and will respond over a
wide band of frequencies--e.g. a piano sounding-board (Fig. 1.8).
Occasionally it is necessary to design a sharply tuned or highly
selective system, such as the resonator tubes below the bars of a
xylophone. But a more common requirement is a flat response,
with either the amplitude or the velocity of the system maintained
substantially constant over a range of frequencies. For example,
in the design of loudspeakers, large diaphragm resonances leading
to the exaggeration of certain frequencies are as far as possible
avoided. In fact it is not possible to make a diaphragm completely free from all resonances and the response curve usually
shows a series of tiny peaks. The violin, on the other hand, uses
the resonances present in the body of the instrument to produce its
characteristic sound.
Throughout the above it has been assumed that the driving
system was more massive or powerful than the driven system. In
cases where the two components in a coujled system are of comparable
mass, the resultant frequency may be intermediate between the
resonant frequencies of the two, approximating more closely to
that of the heavier system--e.g. the reed and the air column of a
wood-wind musical instrument compete for control of the pitch of
the notes, but only in the case of the saxophone (heavy reed) can
the reed take control easily.
Power was defined earlier as the rate of sending out energy, and
a great range of measurements is found to exist. For example, the
maximum power radiated by a large orchestra is about 70 watts,
while for ordinary conversation it is o-0001watts. Whether sounds
cany over long distances from a given source, or quickly become
too faint to hear does not solely depend on this initial power content.
I t is largely a question of directivity. If sounds are to be radiated
efficiently in a given direction, they must be beamed or focused
in that direction at the expense of other directions. (Consider the
old-fashioned speaking-tube.) Radiating equally in all directions
result. in dissipation of the available energy, and a rapid falling off
in volume.
Directivity, in turn, is very much bound up with the size of the
sound source in relation to the wavelength of the sound radiated.
To clear up this important point, let us take the case of (a) a very
small, and (b) a very large source (Fig. 1.g) :1.11.
(a) Vky Small S o u r ~ :in this case, the wavefront takes the form of a
continually expanding sphere-hence the name sphan'cal wavesand the power radiated is spread over a wider area as the distance
increases. Now the intensity is calculated by dividing the total
power by the area of wavefi-ont, and so it is continually diminishing with increasing distance. The surface area of a sphere
increases in proportion to the radius squared, and so we have the
intensity falling off in this ratio, i.e. nim
falls o f with distance
(b) Very Lurge Source: one way of treating this case is to regard
each of the particles of the large source as a point source, when
the effective wavefront of the combination (allowing for expansion at the fiinge) is seen to be a flat area which proceeds outwards
at right angles to the face of the source-hence the name filane
w a s . Thus, from a very large source [email protected] falls off vcry little
with disLnce.
Waves from point sources tend to become plane at great distances
from the source. Conversely, waves from plane sources (e.g. line
Fig. 1.9. oi.g.ammu& rcp'esof sound radhfion from
(a) m y d ,(b) very lmgc sowces
Polar diagram of radiationfrom a
pP;al source
source loudspeaker, Chapter 9) tend to become spherical at great
In the above theoretical treatment, one frequency only was considered. Musical instruments, loudspeakers, etc. are, however,
required to radiate a wide range of ftequencies as evenly as possible.
The criterion as to which of the above modes operates is found to
be the relation between the wavelength of the sounds and the
dimensions of the instrument.
When the length and breadth of the instrument are smaller than
the wavelength, it emits approximately spherical waves with consequent low efficiency radiation in any given direction. At high frequencies, however, where the wavelength is small, the instrument
becomes progressively more directional, and the " carrying power "
is increased, since all the energy is concentrated in one direction
instead of being dispersed over a wide area. An example is given
in Fig. I. 10where the curves show the distribution of sound pressure
around a circular source (a) at low frequency when the diameter
of the source is smaller than the wavelength, and (b) at high
frequency when the diameter iy longer than the wavelength.
All points on curve (a) will have equal'sound pressure, and similarly with curve (b). As an analogy one might say they are similar
to the contour lines of a map, which show points of equal height
above sea level.
When a number of sound waves travel through the same region
of air, an interference pattern is set up in the common region, each
wave continuing on without losing its identity. The resultant
positions taken up by the air particles in the path of both waves are
calculated by adding the displacements due to each of the waves
considered separately.
For example, it is fairly well known (and easily verifiable) that
four regions of comparative silence exist along lines at 45' from a
tuning-fork. The prongs of the tuning-fork form a double source
in anti-phase with cancellations taking place along the directions
A more important example of the combination of sound waves is
the total reflection of a sound back along its original path. The
relative phase of the reflected wave is found to be such as to cancel
the advancing wave at multiples of half-a-wavelength fiom the
reflecting surface (these points are all called nodes), while midway
between the nodes the two waves anive in phase, and reinforcement
takes place. (These points are called antinodes, or loops.)
The interference pattern described above is stationary, and is
called a shnding wnve. It may be noticed in passing that the
amplitude of a standing wave is twice that of the advancing wave
alone, being the sum of the two waves in question. Of course, this
assumes perfect4.e. 100%-reflection.
In general, sounds are
not totally reflected from walls etc., as some of the energy is absorbed
by, or transmitted through, the wall. Thus, the amplitude of the
reflected wave is usually less than that of the advancing wave.
When a standing wave pattern is set up in a studio, small adjustments in the microphone position may cause serious changes in the
quality of sound.
If the dimensions of an object in the path of sound waves are
made smaller than the wavelength of the sound, the amount of
energy reflected is very small. Bending action appears to take
place, and the wave tends to carry on beyond the object as if nothing
had been in the way. Reducing the wavelength of the sound
(increasing the frequency) below the size of the obstacle causes
reflection as usual, and if the wave meets the object at right angles,
a standing wave will result (Fig. 1 . 1 1 ) . This frequency-discriminating property of objects to sounds of different wavelengths is
called the obstacle eflect. Remembering that the range of wavelengths of sound in air extends from about 50 feet down to about I
inch, we see that everyday objects such as tables, chairs, people, and
studio screens, will tend to reflect some frequencies, but not others.
This has an important bearing on the placing of artistes and furniture in the studio if " frequency-discriminated " sound is not to
reach the microphone-i.e. if distortion is to be avoided.
Of the many examples that might be chosen, it will be seen that
acoustic studio screens call for careful use since their width is about
3 ft. This means that, whereas high frequency sounds are effectively
controlled by the screen, sounds of wavelengths greater than 3 ft.
(frequencies below 350 CIS)will " bend round " the screen and not
be reflected. More will be said about this important subject later.
The behaviour of curved reflecting surfaces is also conditioned by
the " obstacle effect
Reflections from a concave surface. for
example, tend to be focused together, but only if the diameter of the
reflector is greater than the wavelength of the incident wave. I t
follows that the parabolic microphone reflectors used sometimes in
outdoor broadcasts and recordings-where they are trained on
distant objects such as the batsmen in cricket broadcasts, song-birds,
etc.-are comparatively ineffective at low frequencies. The two
models available are 3 ft. and 18 in. in diameter, and have " cut-off"
frequencies of about 300 c/s and 600 c/s respectively.
Similarly, convex surfaces, which are used in studios to disperse or
" diffuse " sound waves, arc only effective when the dimensions of
the surface as a whole are greater than the wavelength.
Before describing some of the characteristics and limits of hearing,
a brief word on the construction of the ear may be useful. The
diagram (Fig. 2.1) is purely functional, and not drawn to scale.
The auditory canal leads from the opening in the outer ear, is about
I inch long by f inch wide, and is broadly resonant in the range
2,000 to 6,000 cis. It is closed at the inner end by the ear-drum, an
oval of skin stretched over a bone frame. Attached to the inner
side of the ear-drum is the first of three ossicles or tiny bones. These,
together with their supporting muscles, etc., are contained in the
middie ear cavity which is air-filled, and connected to the throat by
means of the eustuchian tube. This is normally closed, but opens on
swallowing or yawning, to permit air to enter or leave, thus
tqualising the long term pressures on the eardrum.
The ossicles provide some protection against sudden bursts of
sound, and pass on the vibrations of the eardrum to another thin
membrane, the oval window. In this way, the vibrations are conveyed to the fluid in the cochlea, the movement being absorbed by
the round window.
The casing of the cochlea is a bony structure wound in spiral,
form (shown straightened out in the diagram). For the present
purpose, it can be taken to be divided into two galleries by a partition which runs nearly to the apex and which includes the ban'lar
membrane. Reaching out into the cochlear fluid, and responding to
its movements, are a large number of hair cells extending all along
one side of the basilar membrane. These are the endings of nerve
fibres running to the brain and which keep the brain informed of
the vibratory stimulus to the ear-drum.
The brain uses the information sent to it by the ears in two different ways. Firstly, it enables us to hear the sound; that is, to determine its pitch, quality and loudness. Secondly, it performs a
calculating operation on the minute differences between the sounds
received at each ear. The result of this is that we are able to determine the position of these sounds in space. This operation will
be more fully discussed in the chapter on Stereophony (chapter 16).
The human ear will respond to a very wide range of frequencies,
from about 20 to 20,000 CIS--or nearly ten octaves. Sensitivity to
" freauencies is found to deteriorate with acre.
The k n i m u m intensity level that can be heard by a given
observer is not the sameat all frequencies. In most people maximum
sensitivity usually occurs round about 3,000 cis, with a falling off
gradually at lower frequencies, and more steeply at higher frequencies. The graph of minimum audible intensity against frequency is
called the threshold of heating (Fig. 2.2).
As an example of the degree of variation in our hearing abiIity,
the power required to produce an audible sound at 50 cis is about
I,OOO,OOO times that necessary at 3,000 cis.
If the intensity of a given sound is progressively increased, causing
it to get louder and louder, a point is reached where the sensation is
one of feeling or pain rather than hearing. This level does not
depend very much on the frequency of the sound and is called the
threshold of feeling.
As already mentioned in Chapter
I , any two musical notes which
are an octave apart have frequencies in the ratio of 2 : I. If we
nnvcsfm m a g e hearing
Eg.2.2. Equal [email protected]
(Robhon and Dadron, N.P.L.)
write down the frequencies for the octaves of A, we have the following result:
The frequencies are in geometric progression with a common
ratio of 2. I t is for this reason that frequency response graphs etc.
are drawn on a logarithmic scaIe on which equal divisions are
allotted to equal changes in the logarithm of the frequency (powers
of ten) rather than on a linear scale, where the divisions apply to
a fixed number of cycles per second.
The octave is perhaps the most obvious example of a pair of
notes which stand in a recognisable relation to each other. But a
simple ratio of frequencies is found to exist for all the musical intervals which have a simple relationship to the ear. In fact, the order
of conrotuurce (or degree of harmony) of two notes sounding together
is found to depend on the smallness of the numbers used to express
the " frequency ratio
Next most concordant after the octave
are the perfect fifth (3 : I) and the perfect fourth (4 : 3), and so on.
The interval of a second (g :8) is comparatively d u s ~ l ~ (or
n t &cordant), due partly to beats between the harmonics of the two
Our major and minor scales have evolved from the grouping of
notes which provide the largest number of consonant combinations.
For example, the major scale of C, taking the frequency of C as
x, may be laid out as follows:C
This is called the scale of just intottation.
The present system to which keyboard instruments are tuned,
and most music approximates, is called the scale of equal temperament. Here, the octaves are tuned to true pitch, and each octave is
divided into twelve equal divisions called tempered semitoms.
In equal temperament, music can be played equally well in all
keys. There is, however, the serious disadvantage that only the
octaves are in perfect tune. There is evidence that singers, stringplayers, etc., when not accompanied by keyboard in~truments,
tend to depart from equal temperament to something approximating more closely to true, natural intervals.
Our appreciation of loudness is found, generally speaking, to
follow a somewhat different law: if we multiply the intm2y of
sound by 10 the loudness is approximately doubled.
If it be increased again by a further factor ,of 10, that is, roo
times the original intensity, the second change in loudness wauld
appear to the ear to be the same as the first, that is, the sound
will now be about four times as loud. The unit which relates
these changes in intensity is termed the Be1 and so I Be1 is the
amount of change when the intensity is changed by a factor of 10.
Now the change in intensity represented by the two thresholds
at 1,000 cis (see Fig. 2.2) is approximately in the ratio of
1012 : I.
This is twelve factors of
~ , o o o , m , o o o , m : I-i.e.,
ten, and therefore is said to be 12 BeLs. Thus the number of &Is
is the same as the index of the power of 10.
ratio in bels = loglr1,
In practice this is too large a unit and the decibel is used:-
ratio in decibels (dB) =
10 logl,-
Under good listening conditions the minimum change in intensity
that can be detected by the ear as a change of loudness is one
dB =
10 log11
log-l a
antilog 0.1 = 1-2
It will,be seen therefore, that a sound of any intensity must be
changed by 26% before a loudness change can be detected.*
If we refer again to Fig. 2.2, the curves show the intensity in dB
required to produce sounds at equal loudness, at varying loudness
levelp. It will be seen immediately that the intensity varies with
frequency. It is evident then that the decibel is not a suitable
unit to denote a change in loudness level, since for a given dB change
in intensity the loudness is entirely dependent on frequency.
A different unit is therefore used, called the phon, and a sound is
said to have a loudness level of n phons if it sounds equally loud
as a 1,000 c/s tone ndB above the zero level. Zero level is agreed
to be the minimum intensity required just to hear a 1,000 c/s note.
* For a discussion of the decibel as a measure of change in an electrical circuit
see Appendix
How many dB separate the thresholds of hearing and feeling
I ,000 c/s ?
The ratio of the intensities has already been given as
number of dB = 10 log
1 0 :~I ~
Example 2
Sounds in a studio are found to die away to one-millionth of their
original intensity in one second. What is the decay of sound,
expressed in dB ?
The ratio of the intensities is ~,ooo,ooo: I
fall in dB = 10log los
hw'h 3
Given that the logarithm of 80 is approximately I -9, how many
dB more intensity will result from eighty grand pianos playing
together than from one?
The ratio of the intensities is 80 :I
increase in intensity = l o log 80
All musical instruments have this in common: they are designed
to produce one or more pleasing tones. Common to most musical
instruments is a ystem of one or more resonatms, by which is meant
a component possessing a discrete resonant frequency of free vibration. Many structures possess resonant characteristics suitable for
use in musical instruments, including strings, air in pipes, bars,
membranes and electronic oscillators. In a given instrument, the
pitch of the resonator may be k e d , or variable. Some means of
exciting the resonators into vibration is provided, under the control
of the musician.
Examples of instruments possessing the most common methods
of excitation are listed below:
S&gs (4)
strings (46)
Strings (88 notes)
Air column
Air column
Air column
Bow or fingers
Edge tone
Player's lips
There is another part of many musical instruments that should be
mentioned. It is included to improve the radiating efficiency, and
A string, for example,
may be given the general name-radiator.
tends to cut through the air as it vibrates without causing much air
movement. Hence the need for a sounding board in pianos, and
the complicated body of violins, etc.
It follows that the overall quality or timbre of a given instrument
is not decided by the harmonic content of the resonant component
alone (see Chapter I), but is likely to be modified by the selective
characteristics of the radiator. Many instruments possess a region
of pitch-known as the fmmant-in
which tones, fundamental or
harmonic, tend to " sound out ". Violins, for example, derive
much of their characteristic tone from the formant reinforcement
over the range 3,000 to 6,000 CIS. The piano is something of an
exception in that its very heavily damped sounding board responds
c4 3
evenly to all frequencies, and radiates the string tone more or 1faithfully.
Strings are used as the tuneable, or resonant, system in many
musical instruments.
A string is capable of vibrating in several modes simultaneously,
to produce a full range of harmonic frequencies. Fig. 2.3 shows
the sine wave form of the standing wave on a string for the fundamental and the first four harmonics. The fixed ends always being
nodes, the " harmonic number " is seen to be the number of loops
formed along the string.
Only transverse vibrations of the string are considered in music
-witness the unmusical sounds produced by the new violinist who
Fig. 2.3. Hmmonic modes of vibration of strings
fails to bow at right-angle to the strings, and excites dissonant
vibrations along the length of the string.
S t r i n g QPnlity
The effect on quality of the formant has already been mentioned,
and will vary slightly from one instnunent to another. It is not
usually under the control of the player. Great variations in the
tone are possible, however, depending on the point and method of
Bowing close to the bridge (sul ponticello), for example, is sometimes resorted to for especially brilliant quality, rich in harmonics.
A broad bow gives extra mellowness of tone, while using the wooden
back of the bow (col legno) produces a harsh, dry effect. Again,
fairly mellow tone is to be expected when strings are plucked with
a round " instrument ",such as the fingers. Using a sharp plectrum
pulls the string into the shape of a triangle, and much of the energy
is thrown into the highest harmonics, with resultant metallic tone.
As a final example, if a hard, sharp hammer is used, a sharp kink
is given to the string, and, in the extreme case, all harmonics are
formed at equal strength-e.g. a coin dropped on to piano strings.
The felting of piano hammers is calculated to produce a pleasing
proportion of harmonics, and " tinniness " begins when the felt
becomes matted down.
I t is worth noting that the effective hardness of the hammers
increases when the notes are played forcibly in forte passages. This
means that dynamic variations in piano playing are characterised
not only by simple variations in loudness, but by real variations in
quality. Louder passages are more brilliant in tone, due to the
extra production of upper harmonics.
A large group of musical instruments use the resonance of one or
more pipes to produce notes on the scale. As with strings, the
principles involved and the tuneable factors have been known for
very many years. When the air in a cylindrical pipe is excited into
vibration, a standing wave is set up due to reflection and re-reflection between the two ends. The length of the pipe is therefore
related to the wavelength of the standing wave to which a given
pipe will resonate. There is a difference in the relationship, by a
factor of 2, depending on whether both ends of the pipe are open to
the atmosphere, or only one.
End Correction
Because of the change-over from plane waves inside to spherical
waves radiated from the end, the true position of the antinode is
not exactly at the end of a pipe, making the effective length slightly
more than the measured length.
The extra length (or end correction) for harmonic frequencies
is slightly different from that for the fundamental, a factor which
has some bearing on the tone of different instruments. Wide bore
pipes are found to generate fewer harmonics, and sound less
brilliant than narrow pipes, due to the mistuning of partials when
the end correction is taken into account.
Methods of Ecitadon of Air Column
The air column in a wind instrument may be set in vibration in
at least three ways, namely :(a) Edge tones
(b) Reeds
(c) Player's lips
(a) Edge Toms are produced when a stream of air is directed on to a
solid lip or wedge. If this is coupled to a resonant pipe, excitation takes place at the natural frequencies of the pipe, eddies
being struck from alternate sides of the wedge. Typical edge
tone instruments are the flue organ pipe, flute, and recorder.
(b) Reeh, often made from actual reed or bamboo, are used in
some organ pipes and in the clarinet, oboe, etc. Single and
double reeds are possible, the former consisting of a thin wafer
sharpened at the blowing edge, and the latter comprising two
flat pieces bound round a small metal tube at the pipe end. In
each case, the air stream is interrupted by the vibrations of the
reed, so that the associated air column is triggered into resonance.
When a reed is blown in free air. a note is emitted. rich in harmonics. When a resonant air column is coupled in, the natural
frequencies of the column modify the reed quality, and dictate
the pitch of the note.
(c) The Player's L$s are used to excite the air column into vibration
in musical instruments of the brass family. The action resembles
that of a double reed, and the natural rate of interruption of the
air stream will be the resonant frequency of the lips and associated
acoustical system. Usually this will be dictated by the air
column, but some control is possible by changing the lip tension,
and blowing pressure. Typical lip-reed instruments are the
trumpet, trombone, and French horn.
(Fig. 2.4)
The Violin consists of a sound-box of special shape, closed except
for two f-holes. The four strings are stretched over a bridge and
tuned, by adjusting the pegs, to notes a fifth apart, as shown.
Stopping down with the fingers of the left hand gives a fundamental range of four octaves.
The Vjolu has heavier strings, whose open notes are tuned a fifth
lower than those of the violin. Its fundamental range is over three
octaves and its overall length is 24 in. more than that of the
The Violoncello or 'Cello is tuned an octave lower than the viola and
has a fundamental range of over three octaves.
The Doubh Bass or Contrabass has heavy strings tuned in fourths,
as shown. Its fundamental range is about three octaves, and it may
be bowed or plucked (Fig. 2.4).
The Harp consists of 46 strings stretched on a triangular fkame,
one side of which 5 a small sound board. This sound board is a
relatively inefficient radiator, and the harp tone is characterised by
the slow decay of sound. Seven pedals serve to sharpen the pitch
of the notes-the C pedal controls all Cs and so on. The pitch
shifting may be performed in two steps-a sernitone or a full tone.
The fundamental range is 6) octaves, as shown.
(Figs. 2.5 and 2.6)
Th Flute consists of a cylindrical tube which resonates as an open
pipe. It is excited by means of a blowing hole for producing edge
tones and is held in a transverse manner across the player's mouth.
d f
Fig. 2.6. The brassf d y
The blowing hole is situated a little distance from the actual end of
the instrument, which is closed by a stopper, and a special open-end
correction factor applies. The various notes of the scale are produced by uncovering holes along the pipe, thereby shifting the
effective position of the open end. In early flutes it was necessary
to make the holes of a convenient size, and a convenient position for
the fingers. This imposed a limit on the variety and accuracy of
notes. With the introduction of keys, and especially through the
work of Theobald Boehm, great flexibiIity and trueness became
possible. The modern flute has a fundamental range of three
octaves from middle C.
Associated with the flute is a smaller transverse instrument called
the Piccolo, which plays about an octave higher.
Tb Oboe is a double-reed instrument with a conical tube. Its
fundamental wavelength is related to twice its length, and it is
capable of producing the full series of harmonics. The oboe's note
A is often used for tuning purposes in orchestras, but a tuning-fork or
electronic tone source at 40c/s is more reliable. The fundamental
range of the oboe is similar to that of the flutethree octaves from
Associated with the oboe is a larger instrument called the English
Horn, or COYAnglais, whose conical tube ends in a spherical bulb.
I t plays about a fifth lower than the oboe, and has a smaller compass.
The Bassoon is a double-reed instrument with a long conical tube
(93 in. long) doubled back on itself for convenience. Its fundamental range is about three octaves, at a pitch two octaves lower
than the oboe.
The Contra-Bassoon is folded three or four times, and plays one
octave lower than the bassoon.
The Clarinet is a single-reed instrument with a cylindrical tube.
Its fundamental wavelength is associated with four times its length,
and its characteristic tone derives from the preponderance of oddnumbered harmonics produced. The system of keys is more
complicated than in, say, the oboe, since over-blowing raises the
pitch to the third harmonic (the nearest strong overtone), an interval of a twelfth. The fundamental range of the clarinet is over
three octaves from D.
The Bass Clarinet covers a range one octave lower than the
The Saxophone is a single-reed instrument with a conical tube
employing (except in the soprano saxophone) a curved mouth-tube
and upturned bell. The tube is wide compared with the clarinet,
which makes for the more rapid attack, or " speaking " of the saxophone. There are five members of the saxophone familysoprano, alto, tenor, baritone, and bass, of which the first and last
named are rare. Each type of saxophone covers a fundamental
range of about 23 octaves, and the basic pitch falls in fifths from one
type to the next. The fingered notes on the different saxophones are
given the same names, the necessary shift in pitch being conveniently arranged by appropriate transposing from the written
The French Horn consists of a tube about 12 ft long coiled on
itself, and is a narrow cone with a wide bell. The instrument has
a funnel-shaped mouthpiece to which the player applies his lips,
varying their tension to produce the effect of a double reed. I t is
possible to make the air column resonate, and play notes corresponding to about the first fifteen harmonic frequencies. Further
notes, to cover the full chromatic scale, are produced by coupling in
extra lengths of pipe. The earlier method of fitting a choice of
U-shaped " crooks " to correspond to changes of key in the music
gave some extension of the range. The modern system of three
valves which introduce any combination of three extra lengths of
tube has produced a flexible instrument with a fundamental frequency range of over three octaves.
By inserting his hand into the bell of the French horn the player
is able to modify the effective length of the tube, in such a way as
to raise or lower the pitch, or produce muted effects.
T h Trumpet has a cup-shaped mouthpiece and wound tube, 6 A
in length, which is partly cylindrical and partly conical. Three
valves, together with harmonic selection by the player, provide a
chromatic scale over about three octaves. Metal or plastic mutes
may be fitted into the flared bell to reduce the output and give a
change in quality.
T h Cornet looks like a more compact version of the trumpet, and
covers a similar range of notes. Its tone is less brilliant, and it is a
little easier to play.
T h Trombone has a cup-shaped mouthpiece and a U-shaped
cylindrical tube terminating in a flared bell. The tube is about nine
feet long, doubled on itself, and possesses a telescopic section or slide,
giving continuous variation of the resonant length of pipe. Sliding
or glissando effects are therefore possible, and the intonation of
notes depends on the player as in the violin, etc.
There are two sizes of trombone, the tenor, with a fundamental
frequency range of about two and a half octaves, and the bass, with
a range of about three octaves.
Thc Tuba is the largest orchestral brass instrument, and has an
18 ft coiled tube ending in a large bell pointing upwards. Three,
or sometimes four, valves are used to give a fundamental frequency
range of about three octaves.
In all percussion instruments, the vibrations result from the
instrument being struck by hammers, drumsticks, etc. The vibrating systems may be bars, plates, bells, or drumskins. I t is possible
to divide the percussion family into two p u p s , depending on
whether the sounds are of definite pitch or not.
Indefinite Pitch Instramemts
In this category come the triangle, bass drum, snore drum, cymbak,
and gongs. A very wide range of frequencies is generated by this
group, including high frequencies up to and beyond the limit of
Dehite Pitch Instruments
In this category come the xylophone, glockenspiel, cehste, tubular
and other bells, and the kettledrum or timpani. A feature of these
instruments is that the overtones are not usually simple multiples of
the fundamental frequency.
i% Pianoforte is usually listed as a percussion instrument. It
consists of a keyboard, and a system of hammers which strike
steel strings, stretched on a steel frame. A bridge couples the
frame to a sound board which runs the whole length of the instrument. The fundamental range of the usual piano is over seven
octaves-eequency range of 27'5 to 4,186 CIS.
Great variations in intensity are also possible (dynamic range)--so
that this instrument puts the reproducing systems of broadcasting
and recording to a severe test.
Depressing one of the keys flicks a hammer into momentary
contact with the appropriate string(s), and at the same time withdraws the damping pad. When the key is released, the hammer
Fig. 2.7. Some pnmsion irrthments
falls back into place, and the damper bears on the string and silences
Operating the right pedal, or sustaining pedal, removes all the
dampers from the strings, so that the notes decay comparatively
slowly. I t also permits sympathetic vibration of strings which are
harmonics of struck notes, causing considerable reinforcement of
tone. A centre pedal is fitted in many pianos, called the bass
sustaining pedal, which removes the dampers from the bass strings
only. The left or soft pedal reduces the volume of sound in some
way-by reducing the length of stroke of the hammen, or by shifting the hammers so that fewer strings are struck, or by keeping the
dampers in contact with the strings.
The voice mechanism may roughly be described as a double-reed
musical instrument. The vocal chordr, when suitably tensioned,
interrupt the steady flow of air forced between them by the lungs
(Fig. 2.8).
The pitch of the fundamental note produced depends on the
tension, thickness, and length of the vocal chords. All of these are
capable of some variation in speech and singing, the basic length
being greater for men than women. The male speaking voice has a
Fig. 2.8. The human coue
range of about 12 tones, centred on 145 CIS. The female speaking
voice has an average frequency of 230 c/s with a similar compass.
The stream of air, pulsating according to the saw-too'thed interruptions of the vocal chords, emerges into the cavities of the mouth
and nose. I t is the variation of the resonances of these cavities by
manipulation of the lips and tongue which controls the harmonic
Fig. 2.9. Comp(~(sof singing wicu
content of the sounds. Changes in vocal quality, not to mention
emotional quality, correspond to variations in the relative strengths
of partials. Investigations have shown that the production of each
of the various vowel sounds depends on prominent resonances within
two regions of pitch. These resonances, or more correctly bands of
frequencies, are known as formants, and are usually specified by
single frequencies. It must not be forgotten, however, that a
formant frequency is in fact the centre of a wide band of sound.
For example the vowel sound in the word " tone " is produced
when frequencies in the region of 500 c/s and 850 c/s are stressed.
The two formant frequencies for the vowel sound in the word
" soon " are fairly low-400
c/s and 800 CIS-which explains the
difficulty in singing this vowel clearly on high notes.
The fundamental frequency range of most singing voices is about
two octaves, and average values are shown in Fig. 2.9.
IN designing a theatre or concert hall for good acoustics, the
following problems arise :(a) keeping out unwanted noise;
(b) obtaining adequate loudness at all points, without
spots ,';
(c) obtaining good definition and intelligibility.
It will be seen that the last two requirements are in opposition,
(b) calling for a gradual, and (c) for a rapid decay of sounds.
Broadcasting studios have a slightly different requirement, since
directional microphones can be used to pick up sounds.
Television studios present special problems, due to the erection of
a number of settings simultaneously, and to the high level of background noise often encountered.
An audience in a theatre will tolerate a level of extraneous noise
which would be very distracting in a broadcast. This is because
they are able to use both ears to locate the direction of the noise,
and unconsciously ignore it, whereas the microphone does not
discriminate against noise in this way. It follows that soundproofing is more important for studios than ordinary theatres, and
Fig. 3.1 shows the minimum insulation required for speech studios.
Insulation will normally be better than this, and in the case of music
studios will be of the order of 70 dB.
It is usual to distinguish between:
(a) air-born misoentering through doors, windows, or ventilators,
and including traffic noise, conversation and aeroplanes;
(b) structure-borne nohe--due to impacts on the parts of the building,
footfalls, machinery, slamming doors, and nearby studios.
In the case of air-borne noise, the reduction depends on the mass
of the separating structures, and is most difficult to achieve at low
frequencies. The reduction of structure-borne noise depends on
discontinuous construction which forces sound to travel through a
variety of different materials. Some reflection will take place at each
Fig. 3.r. Minimum jgwu
for studio innrlaria
boundary, resulting in more or less rapid attenuation. In this case
high kequency sounds are often well propagated and their attenuation can be a source of considerable difficulty.
Leakage between a Studio and its Cubicle
Poor insulation between a studio and its cubicle may give the
studio manager a wrong impression of the quality actually being
radiated. It frequently results in the impression of more bass than
is really being sent to line.
Another danger, however, is the risk of" howl-back "-when a
sound received by the microphone returns to the studio at an
appreciable level via the cubicle loudspeaker. This condition will
be set up Xthe gain from the microphone to loudspeaker exceeds the
loss through the walls or doors, and the sounds from the monitoring
loudspeaker will leak back into the studio and complete a circuit
to the microphone (Fig. 3.2).
Sound insulation is usually less at low frequencies, as has already
been mentioned, and the improved bass output of modern loudspeakers can give extra trouble. Howl-back will occur at some
frequency for which the path length is such as to return the sound
to the microphone in phase with the original.
An echo may be defined as a single reflection of sound following a
noticeable time interval after the original. Reverberation, on the
other hand, is a property of an enclosed space, and is the name given
to the prolongation of sounds due to many reflections from the
walls, etc.
When a sound commences in a studio, it does not immediately
build up to its full intensity. The source is supplying energy at a
given rate, but the waves are losing a certain percentage of their
energy on each reflection from the boundaries. Equilibrium
intensity is reached when the supply and withdrawal rates are
equal. This " steady state " condition resembles the temperature
of a room when heat from a radiator is being supplied at the same
rate as it is lost through doors, etc. When the sound source ceases
to vibrate, some time elapses before the reverberating energy is
completely absorbed.
The diagram (Fig. 3.3), shows intensities reached by two pulses
of sound under increasingly reverberant conditions. In the last
Fig. 3.2. Circuit round
which "howl-back" mag
drawing, the reverberation period is long, the build-up is considerable, and the intensity reached is high.
For a given sound, the equilibrium intensity, the time required
for build-up, and for dying away, all depend on the " liveliness " or
degree of reverberation in the studio.
Plafc 3. I .
Par1 oy Studio
.llaida Vale, showing mmtbrane absorbers
Plate 3.2. Port of Studio
Swansea, showivg caviy absorbers
This in turn depends on the fraction of sound energy absoibed
by the walls, etc. at each reflection and on the dimensions of the
studio, which affect the time between reflections. W. C. Sabine
was the first to attempt a precise analysis of these effects, varying
one factor at a time. He defined the reverberation tirm of a room as
I -
- 8 -
Fig. 3.3. BuiM-up and &cay m c s for mcbwes wdh (a)
s h t , ( b ) medium, and (c) long reverberation times
the time taken for a sound to fall to one-millionth of its intensity
(through 60 dB). He also carried out two separate sets of experiments. In the first he measured the reverberation time in a
number of bare rooms of different sizes, and found that the reverberation time was directly proportional to the volume.
In the second, he varied the amount of absorbent material in a
given room, and found that the reverberation time was inversely
proportional to the total absorption present.
From this it will be apparent that if all the available surface area
in a studio is considered to be absorbing sound, the reverberation
time will in h c t be proportional to the cube root of the volume,
since as the volume increases, so does the area of absorbing surface.
Only if the total absorption remains constant at all volumes does
the reverberation time increase with the volume.
Opdmum Reverbedon Time
A certain amount of reverberation is desirable in a hall or a
studio if adequate loudness is to be obtained at all points without
straining the musician or speaker. Reflections which follow soon
after the direct sound help to reinforce the volume of tone produced,
hence the practice of using orchestral shells and other bright
surfaces near concert platforms. But too long reverberation results
in overlapping of syllables in speech with a consequent loss of
intelligibility. In music, it is the definition which is lost, and the
ability to distinguish the separate parts.
There is no law which will tell us how much reverberation will
give the most acceptable compromise between these two requirements. But data has been collected with regard to certain halls
judged by competent critics to be musically good. The results
indicate that the reverberation time should rise nearly in proportion
to the volume (more reinforcement being necessary to obtain
adequate loudness) and that longer reverberation times are pre-
Fig. 3.4.
Optimum reverberation tim fm r n l o ~ ~ cof s dzyermt volume
ferred for choral and orchestral music than for speech. This information has been conveniently summarised for music in the form of
a graph (Fig. 3.4). For speech a time of 0-3-0.4 seconds is usually
considered satisfactory.
It might be supposed that, in a broadcasting system, some
allowance might need to be made for the fact that the listener's
room will have a reverberation time which will be added to that of
the studio. However, in practice this is not usually a difficulty, and
may be neglected if the studio reverberation time is long in comparison to that of the listener's room.
Some average reverberation times for various BBC studios and
public buildings are given in Table 3. I .
Table 51
The figures in the right-hand column represent the average reverberation time
between 500 and I,OW CIS, generally to two significant figures. This frequency
range largely determines the subjective liveness. The studio figures allow for a
typicalnumber of performers, while thaw. for concert halls and opera horepresent
nearly full-house audiences.
Maida Vale, Studio I
Maida Vale, Studio 2
Glasgow, Studio I
Birmingham, Studio 6
Concert Hall, BroadcastingHouse
Belfast, Studio I
Farringdon Hall
Swansea Studio
Camden Theatre
Aeolian Studio I
The Parh Studio, London
Cardiff, Charles St.
Studio 6A, Broadcasting House
D i i o n Studios
Talks Studioa
Concert Ha&, Opno Hma, eic.
Royal Albert Hall
St. Andrew's Hall, Glasgow
Royal Festival Hall
Usher Hall, Edinburgh
Free Trade Hall, Mancheater
Colston Hall, Bristal
Concertgebouw, Amsterdam
Covent Garden Theatre
Glvndebourne O ~ e r aHouse
w i p e r ~heatre,-~ay-reuth
King's College Chapel, Cambridge
Sym hony Orchestra
~ m Combinations
Scottish Orche~tra
Midland Light Orchestra
Small Orchestras, Chamber
Northern Ireland Light Orchestra, etc.
Light Music
Light Ente+nment
Light Entertamment
Welsh Orchestra
Live End
Dead End
(approx.) 6.0
The acoustic design of studios is complicated by the fact that no
material will absorb equally at all frequencies. This means that
more than one form of treatment will usually be necessary, and the
standard approach in BBC studios in the past has been to include
porous absorbers and vibrating panel or membrane absorbers for
the absorption of high and low frequencies respectively.
Porous Absorbers
A porous material such as cotton wool is an efficient absorber at
middle and high frequencies. As the pressure fluctuates in the
sound waves, air particles vibrate in the pores of the material, and
dissipate energy in overcoming friction. For porous materials to be
effective sound absorbers they must have a thickness approximately
proportional to the wavelength of the sound to be absorbed. Hence
for practical absorbers the absorption will be less at low frequencies.
This can be improved by spacing the material from the wall, so
Absorption Coe&id
125 CIS 500 CIS
Typical acoustic tile
2. I in. mineral wool with fabric cover
3. As 2, but with I in. air space behind
1 0.16
4. As 3, but with 5% perforated hardboard j
wver in place of fabric
i, z:;:
Plywood 2 in. air space
Plywood air space stuffed with wadding
Hardboard and roofing-felt stuck together.
43 m. a ~ space.
Membrane absorbers
See Fig. 3.6.
4000 CIS
that reflection from the wall surface behind the absorber can take
place, thereby increasing the effective depth.
Table 3.2 shows the variation in absorption coefficient (i.e.
fraction of sound absorbed) for various materials at different frequencies. I t will be seen that a wide range of values is available.
Vibrating Panel Absorbere
Panels of plywood or hardboard supported about 2 inches out
from the wall by battens one foot apart are found to possess natural
frequencies in the range from 80 to 200 CIS. They are set in vibration by sound waves, and will take appreciable energy from the
waves at and near their resonant 'frequency.
If the panel vibrations are damped, this energy is used up in overcoming friction, and the reflected wave is reduced in amplitude,
causing a reduction in reverberation.* In modem studio treatment, very efficient absorption together with pleasing decoration is
achieved using either membrane absorbers, or cavity absorbers.
Membrane Absorbers
These selective low frequency absorbers consist of ordinary
roofing felt stretched over the front of rectangular boxes (Figs.
Fig. 3.5. C a r h u c ~of
membrane absorber
3.5 and 3.6). They have the following advantages over panel
absorbers :( I ) very efficient absorption-up to 100%;
(2) bandwidth is about half an octave-permitting selective treatment where necessary;
(3) absence of coloration due to re-radiation;
(4) they are cheap and easy to construct-and yet applicable to
large and small studios.
The resonant frequency of the unit at which absorption is a
maximum depends almost entirely on the depth of the box.
Most of the absorption is due to the internal friction of the felt,
and a blanket of rock wool may be inserted where additional
* A lightly damped structure tends to continue sounding for a long time, and
re-radiate the sound. This kind oi coloration has sometimes caused trouble in
studics. The "ringing" of lampahades or music-stands is an example of this
damping is required-to widen the band of frequencies absorbed.
When high frequency absorption is also required, the protective
covering is made of canvas or silk.
To prevent sideways vibration of the air, the cavity is partitioned a t spacings less than a wavelength.
3.3.4. Restriction of Absorption at High Frequencic~
High absorption coefficients at high frequencies may well be
undesirable. The furnishings and occupants of the studio tend to
absorb high frequencies and so additional absorption might well
make the studio too dead. The high frequency absorption of the
units described can be restricted by covering them with perforated
hardboard, which will reflect high frequencies.
e.g. 5% holed hardboard reflects above ~ , m o
slotted hardboard reflects above 4,000 c/s
In a particular studio treatment, four depths of resonant unit,
tuned to 62,80,250, and 300 c/s are used in Studio One, Maida Vale
(Plate 3.1). Since this studio was previously bass heavy, little or
no high frequency treatment was included. The projecting units
also serve to break up the sound waves, and give better diffusion.
If this is not required, the units may be recessed, or fitted along the
wall/roof angle.
Cavity Absorber6
These make use of a principle first expounded by Helmholtznamely, the resonating properties of air in a narrow-necked container. The mass of air in the neck of a bottle, for example, will
vibrate against the " spring " of the enclosed air if an " edge tone "
is produced by blowing across the top of the bottle.
Provided the resonance of cavity absorbers is damped, reverberation will be reduced, and different sizes of absorber may be
used throughout the audio frequency band. For example, in
Denmark a particular dance music studio uses three sizes of cavity
absorbers giving a reverberation time of approximately 0.8 seconds
at all frequencies. In this country, cavity absorbers consisting of
rows of cardboard tubes let into plaster boxes have been used
successfully for low frequency absorption. In Swansea Studio One,
line arrays of eight cavities are fixed to the walls to form an unusual decorative effect (Plate 3.2).
When sound waves are reflected and re-reflected between parallel
walls, the amplitude of the resulting standing wave is a maximum
when the walls are a multiple of half a wavelength apart. Thus,
in a rectangular studio, a harmonic series of resonant frequencies
exists related to the length, breadth, and height (Fig. 3.7).
These resonances are called cigmtones, and are undesirable, since
they cause coloration of the studio output. They are most
pronounced in small studios, where the first few harmonics lie
within the band of speech frequencies.
In some Talks studios, the resonances are so marked that they
cannot be eliminated directly by means of absorbers. I t is then
necessary to use a Microphone Correction Unit (see Fig. 4. I 4).
This has the disadvantage that the direct sound also is" corrected "
while the " boominess " is still a distraction to the speaker in the
So far, we have assumed that the sound energy is distributed
evenly in the studio-but this is not usually the case. Standing
waves are set up near walls, sounds are focused in certain spots by
concave surfaces, and frequency distributions are disturbed by
selective reflections from acoustic materials. For example, the
recent attempts at " directed sounds "-using orchestral shells, and
a fan-shaped hall-have tended to give a hardness of tone, presumably due to the strong direct sound and the uneven decay.
A more diffuse sound is obtained by arranging convex or triangular splays or spherical surfaces. There should still be a region of
strong reflection near the orchestra, for best ensemble playing, and
such reflectors should preferably be overhead or at the sides,
parallel to one another.
A live- and dead-end studio, where all the acoustic treatment is
arranged at one end, gives a useful variety of reverberation conditions within one set of four walls--e.g. Studio 6A, Broadcasting House (see also Chapter I 3).
In all except the very largest studios, the average intensity of
reflected sound is roughly eqpal at all points. Thus, for a given
amplifier setting, the indirect sound reaching the microphoneand therefore the loudspeaker-is the same for all microphone
positions; see shaded region at (a) in Fig. 3.8.
Reducing the microphone distance increases the volume of the
direct sound (b), and if we restore the original volume by fading
down on the amplifier (c) we reduce in proportion the amount
of reverberant sound heard by the listener.
In a given studio, then, it is possible to get a range of" apparent
acoustics " by suitable choice of microphone distance. This fact
has many useful applications. It has the disadvantage, however,
that a widely-spread source-such as an orchestra-may present
several aural perspectives at once, unless the layout is carefully
arranged to avoid this.
Also it restricts cast movements in broadcast plays-the increase
in apparent reverberation due to stepping back a few paces may
sound like moving into the middle distance.
For a variety of reasons, a television studio cannot be built on the
same lines as a studio used for sound only. In particular, the
acoustic problems are quite different. Even the smallest television
vo~unt(CUBIC FEET)
Fig. 3.9. Optimum rcderation timefor General A a p a e TV Studios
studios are large by sound standards, in order to accommodate
banks of lights, cameras and crews, microphone booms, special
effects units, and scenery associated with a number of sets (see Plate
3.3). To bring the reverberation time of such a large enclosure
down to the required value for Talks and Drama necessitates heavy
acoustic treatment. It is not unusual for almost the entire wall
area to be covered with glass wool, held in position by perforated
board or chicken wire. Roofing felt units are also employed.
The curve in Fig. 3.9 shows the variation in reverberation time with
volume found to be most satisfactory.
A further reason for making television studios comparatively
" dead " is the need to keep the microphone " out of shot ",which
often makes the working distance greater than that used in a comparable sound broadcast. As actors, variety performers, etc. move
about, they are followed in many cases by &camera, and also the
microphone. A boom is used, carrying a directional microphone
on a telescopic arm. It is the job of the boom operator to swing
the micro~honeto and fro. and to rack it in and out: at the same
time, by & a b l e manipulahon of a hand swivel, keepkg it pointing
at the actors. This calls for a great deal of skill and experience,
and is beset by three difficulties:-
(a) keeping the microphone " out of shot ";
(b) keeping the shadow of the microphone " out of shot ";
(c) relating the aural perspective to the visual perspective.
(a) and (b) are mainly a question of experience, the operator just
avoiding the imaginary line joining the lens and the top edge of the
televised scene. If a complicated lighting arrangement is used,
microphone movement may be severely limited, and the two facilities should be planned together when possible. (c) is a constant
problem, although the ear is extremly tolerant of anomalies of
perspective when the eye is also in use. It is possible to tine up the
correct apparent perspective for ear and eye in a given shot--say a
close-up. If the camera now moves back, or switches lenses, or a
change is made to a more distant camera, the boom operator must
move the microphone above the rising lens boundary, and take up a
position which " sounds " at the new distance of the picture. This
operation is complicated, and not one in which any rules can apply,
especially when one remembers that in the monophonic* system of
sound reproduction distances are not what they seem.
The increasing use of " stemphonic " as a term to describe a two-channe
system so arranged as to produce a spatial distribution of sound haa :'ec*aitated a
new term for singlezhannel recordings and transmkions. The term monaural "
scarcely fills the bill, since most people listen to one loudspeaker with two cars.
The term would be strictly true only of single headphone listening. The tenn
" monophonic ",signifying one sound source (i.e. one budspcaker) would appear
to be more suitable and is earning into general uae.
Before leaving the subject of television studio acoustics, the effect
of the actual sets themselves must be mentioned. Panelled flats
and the like in a room setting will mask the acoustics of the studio
proper. The setting up of marked reflections or standing waves is
even more undesirable when the microphone is moved during the
action, and canvas scenery which is transparent to sound is to be
All the thousands of chemical and everyday substances which
exist have been found to consist of different combinations of a comparatively small number of basic materials known as c h t J . The
gases hydrogen and oxygen, for example, are elements. Ordinary
water is found to be not an element but a combination, in certain
proportions, of these two gases (the ubiquitous H,O). There are
about a hundred different elements, and science has discovered
here a further simplification in the grand scheme of nature. All
Fig. 4.1. &pcsmtution of
atom possessing three tlcclrons
the various elements are made up from combinations of a mere
handful of basic units, which are recognised only by their weight and
electrical charge.
The smallest particle of a given element is called an atom. Atoms
are in the news nowadays, and most people know that the atom is
constructed like a tiny solar system, with a heavy nucleus round
which lighter particles circle in orbits, like planets round the sun
(Fig. 4.1). The difference between one element and another is
simply based on the numbers of planetary and nuclear particles
which go to make up their atoms. The planetary particles are
called electrons and the principal nuclear particles are called protons.
Hydrogen, the simplest element, has atoms in which one electron
circulates round a single proton. The electron is prevented from
flying away from the proton by a force of attraction existing between
them, similar to the force of gravity which keeps the moon in orbit
round the earth. The force which keeps the electron in its orbit
is electrical. The electron and proton are said to carry equal
negative and positive charges of electricity respectively, which
produce a force of attraction between them. Like charges, on the
other hand, are found to repel each other. Normally, the total
numbers of electrons and protons are equal, the electrical charges
cancel, and everyday substances show no overall charge.
In some materials the electrons and protons are held together
very strongly and it is difficult for any electrons to be detached from
their atoms. These materials are called insulators and include glass,
rubber and bakelite.
In other materials, notably carbon and most metals, the force of
attraction between the protons and the electrons in the outer
orbits is small enough for these electrons to be easily detached from
their parent atoms. It is these free electrons moving within the
material that can constitute an electric current. The force needed to
maintain this current is known as the electromotive force (e.m.f.) and
is measured in volts. The e.m.E can be produced by a battery,
which works by electro-chemical action, or by the effect of a
magnetic field, as will be seen later in the chapter.
The flow of current is measured in amperes (or amps), units which
correspond to a certain number of electrons per second passing a
given point in the circuit. This may be compared with measuring
water flow in a pipe in gallons per second. As an indication of the
minute scale of quantities in the atom, the number of electrons per
second corresponding to one amp is :
I t may be noticed that the conventional direction of current flow
is from the positive (red) terminal to the negative (black) terminal,
while the actual direction of the electron stream is opposite to this,
because the electron is a negatively charged particle.
The question now arises-how many amps of current will flow
in a given circuit for the application of a given number of volts?
Circuits vary a great deal, and a thin wire of copper, for example,
offers more " resistance " to current than a thick wire. Ohm
experimented extensively on this phenomenon and found that in a
given circuit the current is proportional to the e.m.f., so that doubling
the volts doubles the amps. This proportionality is called Ohm's
Law, and may be written:-
-vI = R
where V is the e.m.f. in volts, I is the current in amps, and R is a
constant for the circuit. Ohm called this the r~sistance of the
circuit. The unit in which resistance is expressed is the ohm which
may be defined as follows:-A circuit has a resistance of I ohm if an
e.m.f. of I volt produces a current of I ampere. Resistance might
also be defined as the ratio of voltage to current.
Exmple I
The e.m.K applied to a certain circuit is I 2 volts, and the current
is 3 amps. What is the resistance in ohms?
= 4 0 h
Example 2
How many volts must be applied to a circuit of 50 ohms to
produce a current of 3 amps?
V/I= R
= 150 volts
43.2. Variable Resistors
We have seen that the current resulting from a given voltage
depends on the mistance in the circuit. Thus it is possible to use
resistors as circuit components to control the current. These
resistors can be made from special resistance wire, whose resistance
is known to be " so many ohms per foot ",or from blocks containing
powdered carbon.
In Fig. 4.2, for example, the current will be 6 amps (12 + 2), or
4 amps (12 + 3), depending on whether the switch is thrown to
the left contact or the right.
If two resistors are connected in series-i.e., end to end-the same
current will flow through both, and its value will be determined by
I Ir
Fig. 4.3. f+&stms
the sum of the two resistances.
to circuit (b) when
In Fig. 4.3, circuit (a)is equivalent
Now applying Ohm's Law to the series circuit, we have three
voltages :V = I X (R1+R2)
x R,
V, = I x R,
V, = I
and the applied voltage V is seen to be divided into two parts,
called the " voltage drop " across R,and R,. The ratio into which
V has been divided is the ratio of the two resistors.
This is the basic circuit of apotctttialdivider, and if a sliding contact
is provided to produce continuous variation of the voltage tapped
off across A B, we have a potentiometer or fader (see Fig. 4.4).
Resistors in Parallel
If the ends of two resistors are connected together (Fig. 4.5), and
a voltage is applied to both, the current will divide into two parts.
bjl:Fig. 4.4 V d l c rLzictors
Right: Fig. 4.5.
tars in parallel
The cu,rrent.through R, and R , are found from Ohm's Law to be
the current is divided in the ratio
which is the inverse ratio of the resistances.
(a) This is what we should expect-more current will pass through
the smaller of the two resistancesand a " current divider " has
been constructed.
(b) The effective resistance of R, and R, in parallel is not found
from the formula R = R,
R, as in the series case, but &om
A battery is doing work when it drives a current round a circuit,
and the rate of doing work is called the Power. The unit of power
is the watt, and the number of watts being consumed in a circuit
driven by a voltage V is given by the formula :44.
By Ohm's Law, replacing V by IR,
W = I x IR
Similarly by replacing I,
This shows us that increasing either V or I increases the power. ,
Note that although we talk of the power being expended, or used
up, it is a necessary consequence of the Law of Conservation of
Energy that the electrical energy must reappear in some other form
-as heat, light, or motion.
FsPmp11eaon E l 4 d Power
Exam#.t I
An electric fire has a coiled wire element whose resistance is
40 oohm What is the power dissipated when the fire is connected
to zoo-volt mains ?
= 1,000 watts
What current is taken by a roo-watt electric lamp connected
to 240-volt mains?
W = VI
= I00
+ 250
= 0.4
Imagine that the two lines at C in Fig. 4.6 are a pair of metal
plates with an air space between them. Close the switch S.I and
electrons will tend to flow round the left-hand circuit. They will
be attracted to the positive terminal of the battery, and repelled
from the negative terminal. Since no current can flow across the
,air space (air is an insulator), a condition of strain is set up, with
an excess of electrons collected on the lower plate. The plates are
said to form a condenser or capacitor, and if switch S.I is now
opened again, the charge on the two plates is held in suspension.
Closing S.2 completes the right-hand circuit, and electrons will
flow through R until the normal uncharged condition is restored.
The storing ability of a capacitor is called its cabadtame (C), which
is measured in Farads. A common practical unit is the microfarad, pF; I farad = r,ooo,ooo pF. A capacitor has a capacitance
of I farad if a pressure of I volt sets up a charge of 6 x 1o18electrons
(the same number as flow per second in a current of I ampere).
The capacitance of a given pair of plates is found to be directly
proportional to their area, and inversely proportional to the
distance between them. If the insulator between the plates is mica
or paper instead of air, a larger capacitance results. The " capsule " of a condenser microphone is an example of a capacitor, with
the diaphragm forming one plate and the fixed plate of the microphone the other (see Chapter 5).
A charged capacitor, with its positive and negative plates, has
been considered as a means of storing electrical energy. A magnet,
with its North and South poles, stores energyin the formof a magnetic
field. This energy can be put to use, attracting pieces of iron or
steel, just as a charged ,capacitor can drive electrons round a circuit.
The process of magnetising a piece of steel is one of aligning
groups of molecules, each group acting as a tiny magnet, so that they
lie in line hitead of,in a random arrangement. ~ h & groups
molecules are termed domaim. The domains of iron are comparatively easily turned, whereas hard steel, once magnetised,
retains its magnetic qualities for a long time (Fig. 4.7).
When current lis passing along a wire, a magnetic field is
produced, and if the wire is wound into a coil, the magnetic field
is identical to that which surrounds a bar magnet. The polarityi.e., position of the North and South poles-depends on the direction
of flow of the current (Fig. 4.8). The coil in a tape recording head,
for example, is fed with alternating current so that the " polarity "
of the poles is continually being reversed, twice in each cycle.
Now consider the circuit in Fig. 4.9, with a coil L and a resistor
R, in series. When the switch is made, current will begin to flow,
but will not immediately be at the rate decided by R alone, since
the magnetic field is being built up-which requires energy. Once
L is fully " charged ", so to speak, no further energy is required to
keep it in its magnetic state, and I = V/R, as Ohm's Law would
predict. In the same way, when the voltage is removed, the
collapse of the magnetic field tends, for an instant, to produce a
voltage to maintain the current, thus paying back the energy
" borrowed ". In this way a coil of wire resists changes in the
current, and this property is called inductance. The unit of inductance is the henry.
If a core of iron is fitted inside the coil, the magnetic field is much
strengthened, and we have an electromagnet.
The collapse of the magnetic field in the example above can be
thought of as a " motion " of the field relative to the coil, and this
motion produces the voltage in the coil. In similar fashion, if a
Fig. 48. Magdcfiid sunoundiffg an r&
permanent magnet is moved inside a coil of wire, thereby creating a
moving magnetic field, a voltage is induced in the wire so long as
the magnet is moving. A voltage difference will therefore exist at
the terminals of the coil. This is the principle of the simplest form
of electrical current generator.
A [email protected] makes switching at a distance possible. Applying a small
current to the electromagnet in Fig. 4.10 energises it, so that it
attracts the iron lever controlling the switch. Thus the cue lamp
24 oR SOY
A8. 4.10. Rcky uud in cur
li& circuit
circuit is completed, and the lamp will light. By extension of the
relay circuit, there is almost no limit to the distance at which the
circuit will operate. The IOO kw Third Programme transmitter
at Daventry for example, is controlled by relays operated in the
main control room some considerable distance away, and this control
circuit has in f
h been extended in the past so that the transmitter
could be switched on in the London control room.
Buzzers and electric bells work on the same principle, with the
addition that positive action of the attracted lever is made to
break the d a y encrgising current, so that the lever Ealls back, and
the action is repeated for as long as the buzzer switch is d e p r d
(Fii. 41I).
4773.- T
I n a transformer there are two coils of wire wound on a core,
electrically insulated h m each other. In a spacial type of transformer, the auto-transformer, the two windings are connected in
series, but the operation is the same. It is convenient in drawings
Fig. 4 1 1 EQcbicbJI
to show the coils separately. We have seen that at the moment of
switching on or off, the build-up or collapse of the magnetic field
surrounding a coil tends to oppose or reinforce the current.
In a transformer, changes in the magnetic field associated with
current in the pnpnmwy
winding induce an e.1n.f. in the S C C O ~
In Fig.4.12,for example, the current meter in the secondary
winding will give a momentary reading first in one direction and
then in the other, as the switch S.I is closed and then opened.
A special type of transformer with two secondary windings is
sometimes used in studio circuits. I t is called a hybrid t r a m f m ,
and it divides the input into equal and independent halves in such
a manner that no interaction can take place between them. With
a normal transformer, any signal fed into one of the secondary
windings would appear at the other and the hybrid is so designed
that this cannot occur. (See Fig. 7.19.)
When a conductor carrying current is placed in a magnetic field,
there is an interaction between the two magnetic fields-the static
one and the one associated with the electric current. The general
effect of this interaction is to produce a force tending to cause the
conductor to move. The direction of movement is at right angles
to both the current and the magnetic field, and it is reversed if
either the current or field is reversed.
This principle is employed in the moving-coil loudspeaker.
Electric currents which correspond to the original sound waves in
frequency and amplitude are passed through a coil of wire. The
coil is suspended in the field of a permanent magnet, specially
shaped to produce a radial field, and is accordingly set in to and
fro movement. These vibrations are taken up by the paper cone to
which the coil is attached and sounds are radiated into the surrounding air. (See Loudspeakers, Chapter 6.)
The process of converting electrical energy into sound energy
just described for a loudspeaker is reversed in the moving-coil
microphone. Here a coil is suspended in the field of a magnet, as
before, and attached to a light diaphragm capable of vibrating in
response to sound waves. When the diaphragm-and therefore
the coil-moves to and h.an e.m.f. tends to be induced in the
coil. If a circuit is comple;ed, perhaps to an amplifier, a current
is produced comsponding to the original sound vibrations. (See
Microphones, Chapter 5.)
The electric currents discussed at the beginning of
this chapter
flowed in a certain direction round the circuit (plus to minus) as
driven by a battery or similar electromotive 'force. These are
called direct currents (d.c.). In broadcasting we are more often
concerned with currents which are continually reversing their
direction. These are called alternating currents (a.c.), and
the number of double reversals (cycles) per second is called the
fiepuny. This lines up with our treatment of sound-wave theory.
Free electrons in a wire carrying I ,000 c/s tone currents are performing simple harmonic motions identical to those performed by
air particles in the sound wave from a I ,000 c/s tuning sound.
The moving-coil microphone described in section 4.7.5. is
seen to be a generator of alternating currents. Drivm by the sound
wave, the coil moves in the field of the magnet, first in one direction
and then in the other, giving rise to an e.m.f. which is continually
reversing. Imagine that the graph of the fork vibrations is a sine
wave. This wave form will apply to the coil vibrations also and
therefore to the e.m.f., the voltage performing 1,000 swings positive
and negative per second.
It is now proposed to summarise the behaviour of a.c. in circuits
containing resistance, capacitance, and inductance.
48.1. AC. and Resistance
Fig. 4.13(a) shows an ax. generator supplying current to a
circuit consisting of a resistance R.
The current is given by Ohm's Law ( V = ZR) as with d.c., and
is the same at all frequencies. Voltage and current reach their
maximum values together-i.e. they are in phase. To calculate the
power in the circuit, peak or maximum values are not used: instead
an average known as the root mean square value (r.m.s.) is
Pig. 4.13. [email protected] opfmition o! rhsjlow of ax. of (a) rrcirlance, (6) capadmu,
and (c) inductanC6
taken and this is found to be approximately 0-7 times the a.c. peak.
Thus, an ax. power supply rated at 230 volts swings between 325
volts and - 325 volts.
shows a.c. applied to a capacitor. The alternating
e.m.f. causes an alternating charge to appear on the capacitor,
with electrons accumulating first on one plate, and then on the
other. Thus, although we have seen that d.c. will not flow through
a capacitor, a.c. can be said to flow in and out of it. The magnitude
of this current (number of electrons per second) is found to increase
with frequency, and also with the capacitance. Thus the opposition of a capacitor to current flow is imersely pro#ortional to both
frc4uncy and capacitance.
6 3 . A.CdLrchrctrncG
Fig. 4.13(c) shows a.c. applied to an inductor. We have seen that
inductance opposes change in the current and might expect the
opposition to increase with fkequency. Such is indeed found to be
the case. Increasing the size of the inductor also increases the
opposition to a.c., and so this opposition in the case of an inductor is
propwtional to both fiequcncy and induckrnce.
The property of opposition to alternating current flow in the
capacitor and the inductance is known as reactance (X), in order to
distinguish it from the similar property of a resistor known as
resistance (R). In most cases, of course, in a given component,
reactance and resistance will be present together-a coil of wire
forming an inductance must have a small resistance-and when
both are present, the combined effect of both resistance and
reactance is known as impcdanu (3(e.g. the impedance of a loudspeaker is often quoted as 15 ohms where the measured resistance
may be only of the order of 5 ohms: the difference is made up of
reactance due to the inductance present).
If an alternating current is applied to the primary winding of a
transformer, each r e v d of direction of the current will cause a
change in the direction of the magnetic field in the core, and these
changes in their turn will cause an alternating current in the
secondary winding of the transformer. The magnitude of the
voltage measured across the secondary winding, due to this secondary current, depends upon the turns ratio between the primary and
secondary. It is therefore possible, by suitably adjusting this
ratio during manufacture, to arrange for the secondary voltage to be
stepped up or down as required in relation to the voltage applied
to the primary.
It might appear that when a secondary voltage higher than the
primary voltage is required, " something for nothing " is obtained.
This, of course, is impossible since the flower taken &om a transformer cannot be more than the power put in and so, when the voltage
is increased, the current in the secondary winding is proportionately
where TIand Tsare the number of primary and secondary turns.
A transformer may have several secondary windings where
different ratios and voltages are required.
The transformer can also be used to convert from one impedance
to another and the impedance ratio
All sources of voltage, such as batteries, microphones and amplifiers possess internal resistance, and since reactance may be present
also, this is given the general name output imbedance. It is the
impedance measured " looking back " from the output terminals.
It is not proposed to discuss this difficult subject at any length, but
one important aspect deserves mention-namely, that a source delivers maximum power into a circuit when the impedance in the
circuit is equal to the output impedance of the source. The source
and "load " are then said to be "matched "--compare the use of a
matching transformer in microphone amplifiers to step up the low
output impedance of the microphone to the considerably higher
impedance of the first valve.
In devices which have only limited linearity, for instance, valve
amplifiers, this principle has to be modified to " maximum undistorted power ",and in such cases the source and the load impedance may be very far from equal.
Numerous uses are made of the frequency discriminating properties of capacitors and inducto-for
example, in the equalisation
of Post Office lines and gramophone pick-ups, in the smoothing
circuits of mains units, and the tuning of radio receivers. Suppression of low frequencies is achieved by inserting capacitance in series.
High frequency attenuation results from inductance in series. The
familiar Microphone Correction Unit used to compensate for the
100 1.000
rise in bass for close speech on ribbon microphones has a circuit
and response graph as shown in Fig. 4.14.
Since the reactance of an inductor increases with frequency
while that of a capacitor decreases, there is one frequency for any
combination of the two components at which their reactances are
the same.
This is called the resonantfrequcncy and the circuit thus formed is
called a tuned circuit. Resonance in this case is analogous to acoustic
resonance mentioned in an earlier chapter, indeed, acoustic calculations can be performed in analogue form using electrical symbols.
It is possible to connect L and C either in series, with respect to the
generator, or in parallel--see Fig. 4.15. Without going deeply into
the theory, it is found that the reactances cancel each other in the
series circuit, so that the current rises to a maximum at resonance.
In the parallel circuit, the current drawn from the source falls to a
minimum at resonance.
The sharpness of resonance depends on the amount of resistance
present. If R is high, the circuit is said to be broadly tuned. Pursuing the analogy between electrical and mechanical resonance
(see Chapter I ) , resistance in the former corresponds to fiiction in
the latter.
Tuned circuits find application in radio receivers and transmitten, oscillators, and some equaliser circuits.
A thennionic valve consists of an evacuated glass or metal tube
in which are a number of metal " electrodes " which connect to
pins or terminals on the outside. A low voltage, usually about 4
or 6 volts, is connected to two of the pins, so that current will flow
through the first of the electrodes, known as the heater o r j h t .
The heating effect of the current causes electrons to be emitted
from a special coating which has been applied either to the filament
itself (in directly heated valves) or to a separate electrode called the
c a t h d (in indirectly heated valves). Provision of one more
Fig. 4.15. tkmmw in
(a) &, ond (6) in
pmallrl rircuicr
electrode, known as the anodd, produces the simplest type of valve
-a diode (Fig. 4. I 6).
If an alternating voltage is applied between anode and cathode,
pulses of emitted electrons will flow to the anode during its positive
half-cycles, but no electrons can flow from anode to cathode in the
remaining half-cycles. The diode valve is thus seen to be a oneway device, and one of its principal uses is to convert a.c. into d.c.
Such a rectifier circuit is found in mains units used with condenser
The triode valve is constructed in the same way as the diode, with
the addition of a further electrode. This consists of a fine wire
Abow: Fw.4.16.Lk% d u e .
Right: Fw.4.17. Triode o a k
mesh, called the cahol grid, placed between the cathode and the
anode (Fig. 4. I 7).
The stream of electrons fiom cathode to anode passes through
the grid, which therefore exerts an influence on the current flow.
For a given positive voltage on the anode, and " bias " voltage on
the grid, a certain value of current will flow through the valve.
Because of the grid's closer proximity to the cathode, comparatively
small changes in grid voltage will cause large fluctuations in anode
current. This is the principle of the triode amplifier. Applying an
a.c. voltage between grid and cathode swings the anode current
up and down in such a way as to produce larger drops in voltage
through the load R, across which appears an amplified version of
the input.
There is a danger in programme circuits of overloading amplifiers. If the swings in grid voltage exceed a certain value, the
anode current swings will no longer follow this &ithfily (remember
that the valve is a one-way device) and serious distortion may
Many more complicated types of valve are used in different
applications, their names giving a clue to the number of electrodes
they have-for
instance, tetrodes, pentodes and hcxodes, have 4 5
and 6 electrodes respectively.
Since the war a new electronic device which may eventually
replace the valve for many applications has been developed to the
stage where it is now in use in great quantities. This is the transistor,
and in several ways it can be said to be similar to the thermionic
valve. The common type of transistor is a three-electrode device
operating in a very similar manner, at least as far as its external
performance is concerned, to a triode.
In its simplest form, the point-contact tramistor consists of a
base of a semiconductor material, such as impure germanium, and
two point contacts, each like the " cat's whisker " in the early
crystal set. These are known respectively as the emitter and the
collector. I n more recent fonns of the transistor the rather unstable
point contacts have been replaced by other pieces of germanium,
the effect of the point contact taking place at the junctions between
the three pieces. This form, the jwtction tr&tor, has now almost
entirely superseded the point-contact variety and is the type most
likely to be found in audio-frequency equipment.
Comparison between the Transistor and the Triode Valve
Both the normal junction transistor and triode valve are threeelectrode devices, and in this sense are similar. However, for the
valve to operate, the electrodes have to be mounted in an evacuated
envelope, usually made of glass, and the vacuum in this envelope
has to be maintained as thoroughly as possible. The valve operates
by virtue of the flow of electrons &om cathode to anode, the rate
of this flow being varied by voltages applied to the control grid (see
section 4.12). In the case of the transistor no vacuum is necessary,
the device functioning at normal atmospheric pressure. The
current in this case, instead of flowing through a vacuum is flowing
through solid material, and in consequence, the device can be made
smaller, stronger and virtually non-microphonic. No heater is
necessary as in the valve, and this means that the likelihood of hum
induction from a.c. heater supplies is removed. Indeed, troubles
of this sort are less likely to occur for various reasons and the device
is far less subject to deterioration and failure than the thermionic
valve. As mentioned above, in the valve the current is carried by
negative electrons; in the transistor the current is carried by positive carriers as well. These positive carriers are referred to as h o b .
This name arises since they actually are holes in the atomic structure
of the material. (They might be said to be similar to the bubble
in a spirit level.) The hole is positively charged and is exactly
equal in charge to the negative electron. Since the hole and the
electron are of opposite sign they will tend to attract each other and
the electron may eventually fill the hole. No more current can
then flow so far as this particular pair is concerned.
Transistor material can be produced having any desired number
of holes or electrons, the relative proportions of these depending on
the impurities present. During the manufacture of the transistor
the germanium, or other semiconductor material, is very carefully
purified and then, by the addition of minute amounts of selected
impurity, the required proportion of holes and electrons can be
introduced. When there are more holes than electrons the semiCOLLECTOR
conductor is known as p-type, p standing for positive. When the
electrons predominate the material is n-type, n standing for negative.
There are two types of three-electrode transistor, p-n-p and
n-p-n, depending on the type of semiconductor material used for
the emitter, base and collector respectively. We will consider
first the n-p-n type.
4x41. The N-P-N T r ~ ~ s i s t o(Fig.
r 4.18)
In the n-p-n transistor the emitter is made of n-type material and
so negative electrons are the predominating carrier. The emitter
corresponds roughly to the cathode of a valve, and when the base,
corresponding roughly to the grid, is biased positively with respect
to the emitter, electrons can flow from the emitter into the bast. If
the collector of the transistor is now biased positively to the base the
excess of electrons in the base will be absorbed by it. I t will
be seen that the collector corresponds to the anode of a valve. Now
the base is ptype material and so not all the electrons fimm the
emitter will pass on to the collector; some will be absorbed by the
base, being combined with its " holes ". If this were to happen
indefinitely the base would cease to be p-type material and would
not have any charge at all. The charge is maintained by a flow
of current from the source of base bias. Varying this current, for
example by means of a superimposed programme current, varies
the voltage across the baselemitterjunction and so varies the current
taken by the collector.
4x4.~. P-NP Transistor (Fig. 4 19)
This is the more common type of transistor, and in this case the
emitter and collector are of ptype semi-conductor and the base of
n-type. The operation is similar to that described in the n-p-n
transistor, but instead of the movement being of electrons, it is a
movement of holes, and the base in consequence needs to be biased
negatively with respect to the emitter, and the collector negatively
in respect to the base.
The circuit of a single transistor amplifier is shown in its most
elementary form in Fig. 4.20, and along side it the comparative
circuit for a triode valve. It will be seen that in many respects
Fig. 4.20.
Compmison of h.Msistor and !ria& circuits
they are similar but it must be noted that in the case of the transistor
a variation of input current is required to produce an amplified
variation in the current through the collector load, whereas in the
case of the valve a voltage variation at the grid is necessary to produce a current variation in the anode load.
Extremely small size and power consumption making the
construction of highly compact equipment possible.
(2) Almost indefinite life under normal conditions.
(3) Freedom h m a.c. induced hum, no heater supply being
(4) Low operating voltages mean that portable equipment can
easily be run from dry batteries, and mains operated equipment
can be much safer than might be the case with valve equipment.
The characteristics of the transistor vary considerably with
changes in temperature. These variations can fortunately be
minimid by suitable circuit design.
(2) In general the level of self-generated noise, usually hiss, in audiofrequency amplifim using transistors is higher than that in
similar equipment using valves. However, circuit design again
can reduce this to an acceptable value.
Transistorised equipment has so far not come into general use in
the BBC, but certain pieces of equipment have been produced in
transistorised versions. The E.M.I. midget recorder (see Chapter
10) is now produced in transistorised form, and numbers of these
are in service. A series of transistor amplifiers is now being produced and these will be incorporated in new studio equipment. The
amplifier AM 911 is a transistorised amplifier for outside broadcast
A microphone is a device for converting sound energy into
electrical energy. Amongst the desirable characteristics required
of a microphone are:
The conversion should be equally efficient at all frequencies in
the working range-i.e., the waveforms of the sound input and
electrical output should be substantially the same.
The [email protected] Icml should be as high as possible in order that it
may overcome unwanted random electrical disturbances known
as noise, always present in amplifiers.
(3) The polar charactmirtic of the microphone should be the same at
all frequencies in the working range.
(4) The frequmcy respom of the microphone should be uniform
throughout the working range.
(5) Unwanted electrical noise generated by the microphone itself
should be low in relation to the wanted signal.
There are two ways of deriving the force which sets the moving
parts of a microphone into vibration. They are called firssure
o#eration and fiessure-gradient ofitration. Sometimes both types of
operation will be used in a single microphone.
Any microphone whose diaphragm is open to the air on one
side only is said to be prwsure optraicd, and may be thought of as
similar to a quick-acting barometer. The magnitude of the force
on the diaphragm depends on its area and the instantaneous pressure
of the sound wave.
If any sort of cavity is formed in front of the diaphragm, the
pressure and therefore the frequency characteristic tends to show
a peak due to its resonance. Also, owing to the obstacle effect
of (a) ob&k
Pig. 5.1. .
5ant of dulphrugm on h fre-
and (b) rcsmrmrra ofcavity in
rupaue of a p~ssu~e-o&watui
(see Chapter I) an increase tends to result at high fkequencieswhere the microphone is a reflector of sound waves. This highfrequency rise due to the obstacle effect is used by designers of
pressure microphones to compensate for losses in high frequencies
which might otherwise occur. In some early pressure microphones
it gave rise to a considerable increase in response to high frequencies,
and gave these microphones a characteristic " toppiness
When both sides of the diaphragm are open to the &-as
in many
ribbon microphones-sound waves reach both sides, and the
resultant force is due to difference in pressure at the two points.
The magnitude of this force depends on the phase difference
between the instantaneous pressures, and this in turn will depend on
the path difference D and the wavelength. In a given microphone,
D is a fixed distance (about 13 inches in a typical microphone), so
that the phase difference due to travelling this extra distance is
small at long wavelengths and large at short wavelengths.
It follows that the actual force derived in a given P.G. microphone
increases with frequency. The designer must arrange that this
increasing force nevertheless results in a uniform output.
When a microphone is placed close to a source of spherical waves,
such as the human voice, the inverse square law-whereby the
Fig. 5.3. Zwease in bass fw ~Ioscworking in P.G. @ation
sound intensity falls off rapidly with distance-must be taken into
account. This has no serious consequence in pressure operation,
but in P.G. microphones it means that the pressure difference
between front and back due to phase change is augmented by a
difference in actual intensity. This is of the greatest significance
at low frequencies, since the phase difference is so small. The
net effect is summarised in Fig. 5.3-namely, exaggeration of bass
frequencies, getting worse as the microphone distance is further
reduced. The minimum distance should be about
for special effects.
ft, except
The fiolQI diagram of a microphone (or loudspeaker) is a graph of
the relative voltage output for sounds aniving a t different angles.
It is usual to measure angles from the microphone axis, and to draw
a separate polar diagram for various frequencies.
The three basic diagrams met in practice are:-
(a) circlmmnidirectional (associated with pressure operation);
(b) figuresf-eight-two-sided
(associated with pressure-gradient
(c) cardioid-single-sided (associated with pressure and pressuregradient operation combined).
Intermediate characteristics can also be obtained by combining
pressure and pressure-gradient operation in different proportions.
All pressure-operated microphones have a similar set of polar
diagrams-omni-directional at low frequencies, and tending more
to one-sided response at high frequencies. Such differences as
Fii. 5.4 P o b diagram fm flu~ ~ - 0 p minoph0n.s
n d
occur are due to the size and shape of the microphone, and not
to the particular method of generating the voltage.
When the microphone is small compared with the wavelengthsounds up to about ],om cis-no reflection takes place, sounds
arrive at the diaphragm equally from all angles, and the polar
diagram is a circle (actually a sphere, if three dimensions are
At higher frequencies, the pattern changes, due to a combination
of two effects, firstly the obstacle efect mentioned earlier, and secondly
the fact that at oblique angles a sound will arrive a t different parts
of the diaphragm at different times. When the frequency is high,
and the wavelength shorter than the diameter of the diaphragm,
these arrivals may be out of phase, causing partial cancellation.
The overall result for a micro~honeabout three inches in diameter
is to give an acceptance angle with good frequency response of
about 60" and progressive loss of high frequencies at angles greater
than this (Fig. 5.4).
The total indirect sound picked up by a pressure operated microphone tends for the above reasons to contain less top than middle
and bass. The result of this is that in reverberant conditions the
sound produced appears to be lacking in high frequencies, unless a
close technique can be used.
Modern pressure operated microphones are exceedingly small,
some having diaphragm diameters as little as half an inch, and in
these the effect is quite small. It can be further minimised by
fitting a small baffle plate or " biscuit" to reduce the high frequency response on the fiont axis.
Preemue-Gradimt Microphones
The force in a pressure-gradient microphone due to phase
difference a t front and back has been shown to depend on the
Rg. 5.5. Polm diagram fw pracrun-gradient m i m e , showing (a) nduction in
efcctiw path rCngth at oblique angles and (b) truo roo0 ''lioc" angles, and two 80•‹ "dead"
extra path length. This in turn depends on the angle of incidence,
as is shown by the diagram, being a maximum at o0 and 180•‹, and
W n g to zero at go0 and 270'. Thus the microphone is equally
sensitive on both faces but is " dead" at the sides (Fig. 5.5).
For an angle of incidence 8 the extra path length is not FB(= D),
but AB which becomes smaller as the angle increases. Mathematically-inclined readers may notice that
cos 9
and the force, and therefore the output, will be proportional to the
cosine of the angle of incidence.
By symmetry in the other three quadrants, we arrive at the
familiar figure-of-eight diagram.
A heart-shaped (cardioid) polar diagram has become popular
where one-sided pick-up over a wide angle is required-for discussions, choirs, sections of an orchestra etc. It may be obtained
by combining the characteristicsof an omni-directionalelement and
a pressure-gradient element whose maximum outputs are equal.
This is shown diagrammatically in Fig. 5.6. The two elements
are taken to be in phirase to the left of the diagram, so that their
outputs reinforce each other. The phase reversal which occurs
on sounds arriving at the back of the figureof-eight element causes
the outputs of the two elements to be in opposition, so that complete
cancellation takes place at 180•‹,and the combined output falls to
zero. Unfortunately, the polar diagram of the cardioid microphone which is oomtructtd fkom two separate element. will t a d
to vary with frequency. This is partly due to the difficulty in
constructing a pressure operated element which is omni-directional
at all frequencies. Also, it is necessary for the two elements to
possess identical response curves, which is difficult to arrange.
A better maintained cardioid response is possible when a single
element can be made to combine the characteristics of pressure and
pressure-gradient operation. Electrostatic microphones having
cardioid response working on this principle have been used in
broadcasting since about 1935. Cardioid ribbon microphones,
which operate by partially enclosing the rear of the ribbon to give
a combination of pressure and pressure-gradient characteristic, have
been available since the war. More recently, since about 1955,
the principle has been applied to moving-coil microphones, although,
as in the case of electrostatic microphones it was not enclosing that
was necessary but opening a path to the rear of the diaphragm.
A general description of the process will be given by reference
to Fig. 5.7. In this the electrostatic cardioid microphone capsule
is described, but the same principles apply to the other versions.
Pressure-gradient operation is introduced by boring a pattern of
holes through the back plate of the microphone, so that sounds may
act on the back as well as the front of the diaphragm. The atm
path length in this case (on which the force on the diaphragm will
depend) is made up of two parts:D-measured on the outsidcwhich will vary according to the
direction of the sound source (c.f. pressure-gradient operation) and
d-measured through the back plate-which is independent of
the direction of the sound source (c.f. pressure operation).
The overall result is a cardioid-shaped polar diagram, since the
extra path travelled by sounds to reach the back of the diaphragm
(comparedwith the front) is (D d)d, and 0 respectively, for sounds
at angles oo, go0, and 180". See Fig. 5.7 (b), (c), and (d).
The actual length of d is likely to be rather less than D, but by
careful design of the back plate its effective length can be increased
until d and D are equal, so that complete cancellation of sounds
from 180" is achieved. This increase in the effective length of d
can be obtained by making the back plate from two discs with
the pattern of holes not quite opposite each other. Rotating one
disc in relation to the other during the manufacturing process allows
the necessary adjustment to be made.
Since a cardioid microphone can be said to be a combition of
pressure and pressure-gradient microphones, there will be a slight
bass rise under close working conditions, due to the pressuregradient part of the device. This will not be so serious as that which
occurs when pressure-gradient operation is used alone.
Variable Polar Diagram Microphones
By using two of the cardioid microphone capsules described
above, mounted back to back, it is possible to produce a number of
polar responses from the combined pair. This is achieved by the
addition of the output of the two capsules in differing amounts
either in or out of phase. The polar characteristics available by
this means can vary from omni-directional through cardioid to
Electrostatic microphones of this type are in practice constructed
with two diaphragms, one on either side of a common back plate
(see A.K.G. microphone type C. 12).
There are at least five methods of deriving an electrical output
from the vibrations of the diaphragm. The electrical action in each
of these cases will be dealt with briefly before giving descriptions
of microphones which are in current use.
Variations in pressure in the sound waves cause the diaphragm
to vibrate. The resultant alternating pressure on the carbon
granules compresses and releases them, causing the resistance
between the terminals to alternate about its mean value. This
Fig. 5.8. lh carbon
imposes an alternating component on the steady current drawn
from the battery. Using a transformer, the a.c. component may be
tapped off as output (Fig. 5.8).
Advantages of the carbon microphone are its robust construction
and large output. These make it suitable for use in the mouthpiece
of telephones. Disadvantages are that a battery supply is needed,
that it generates a high amount of electrical noise, that the granules
tend to " pack " or stick together, and that the output volume and
dictional pattern tend to vary with frequency.
This microphone, sometimes called the dynamic microphone,
works on the electromagnetic principle, and was briefly discussed
in Chapter 4. As the diaphragm vibrates in sympathy with the
sound waves, the coil which is fixed to it is made to vibrate in the
field of a strong permanent magnet. The resultant induced voltage
will be a t the frequency of the sound, and its strength will vary in
accordance with that of the sound waves. The ends of the coil
are usually connected to a pair of insulated terminals. The
impedance of the coil is low-usually of the order of 20 or 30
ohms, so that a " matching " transformer is necessary to step it up
to 300 ohms when used with BBC amplifiers (Fig. 5.9).
Moving-coil microphones are robust, need no special amplifiers,
and can give good quality. They are therefore suitable for outside
broadcasts and as hand microphones. Most moving-coil microphones are pressure operated, and so are nominally omni-directional.
However, some are rather large in size and variation in polar
diagram may set in as low as 1,000 cis. Moving-coil microphones
movingcml m t c 1 0 p h
are also produced having a cardioid polar diagram as previously
Ribbon Microphone
This microphone also works on the electromagnetic principle.
A ribbon of metal foil is suspended between special pole-pieces
attached to a strong permanent magnet, and combines the functions
of diaphragm and moving conductor. The ribbon is corrugated,
and loosely tensioned between insulated clamps. Nearly all ribbon
Fig. 5.10. fi ribbon
microphones in the BBC are of the pressure-gradient type, being
open to the air on both sides, and having a figuresf-eight polar
diagram. Ribbon microphones can, however, be produced having
pressure characteristics and also a combination of pressure and
pressure gradient, so that a number of polar responses can be
obtained from this type of operation. The impedance of the
ribbon is low-a fraction of I ohm-so that a matching transformer
has to be built into the microphone to step up the impedance to
300 ohms (Fig. 5.10).
Advantages of the ribbon microphone are the excellent quality
that is possible, and the fact that the figure-of-eight polar diagram
is maintained at all frequencies. It is therefore useful as a general
purpose studio microphone-for talks, drama and music.
Its disadvantages are that it is very susceptible to wind noise
and vibration, and cannot be used close to a sound source because
of the inherent bass rise under these conditions.
5.5.4 Electrostadc Microphone (Condeaeer Microphone or Capcitor Microphone)*
This microphone is really a capacitor, with one plate fixed and
the other a flexible diaphragm. As the diaphragm vibrates, the
capacitance varies about its mean value. The resistance R (see
llu skctrattdic
Fig. 5.1 1) is high in value so that the charge on the microphone may
be regarded as constant. Under these conditions an alternating
is produced proportional to the fluctuating capacitance.
The microphone capacitance is very small, and corresponds to a
very high impedance. It is therefore necessary to mount an
amplifier at the microphone head, since even a short length of lead
might cause a serious loss in output. The amplifier is not primarily
used to step up the voltage, but to match the higher impedance of
A condenser is now called a capacitor in electronic practice; the term " conbv " electrostatic
denser micro~hone" is therefore inconsistent and is being
- re~laced
microphone " or " capacitor microphone ".
the microphone to the 300 ohms impedance of the studio circuit.
The coupling capacitor C isolates the amplifier &om the battery
(about IOO volts). For good sensitivity, the spacing between &aphragm and back plate is made very small, and the latter is slotted.
A vent hole is provided to equalise slow changes in pressuresimilar to the way in which the eustachian tube in the human
ear opens periodically to equalise inside and outside pressures.
Advantages of the electrostatic microphone are that high quality
is possible and that in some modern types a series of us& directional
patterns may be achieved in a single unit. A disadvantage is that
a head amplifier is necessary, with its associated power supplies and
multi-cored cables.
Crystal Microphone
Some materials, when subjected to mechanical strain, produce
an electrical voltage across opposite faces. This is known as the
piezo-electric effect and is made use of in the crystal microphone. A
sandwich made of two slices from a crystal of Rochelle salt is arranged
so that the sound causes it to bend or twist, thereby producing a
voltage corresponding to the sound input. The crystal movement
Fig. 5.r z
T h nystul microphone
can either be accomplished by means of a diaphragm attached to
the crystal, or by the sound impinging on the crystal itself. In the
drawing a diaphragm is attached to one corner of a crystal assembly,
while the other three corners are fixed (Fig. 5.12).
Advantages of the crystal microphone are that high output is
possible, and the bulk can be reduced to give omnidirectional
characteristics at all frequencies. A disadvantage is that the high
internal impedance makes it necessary to use low capacity cable
to connect the microphone to the grid of the amplifier valve, if
high frequency losses are to be avoided. This limits the length of
cable that can be used. All crystal microphones normally met are
pressure operated, although some pressure-gradient types are
The last few years have seen a great deal of activity in the field
of microphone design. The introduction of new types to BBC
studios is necessarily gradual, and old and new types are therefore
likely to be used side by side for some time to come. The descriptive notes which follow are intended to help in identifying the
different types and in choosing the best from those available for
different applications. The order of treatment will be the same as
in the previous section. Advice on the positioning of microphones
is contained in Chapter I 3.
Carbon Miupphones
No carbon microphones are used in BBC studios.
A series of microphones of this type has been manuf5ctured by
Standard Telephones and Cables Limited, and will be given the
S.T. & C. type numbers for easy reference (Plate 5.1).
( a ) S. I.&' C.
y I 7C (obsolctc)
On account of its rugged construction, this microphone has been
widely used on a hand-grip for interviews, etc.
It is common practice to speak across rather than directly into the
diaphragm to avoid excessive high frequency response. The
terminals require careful tightening as the cable tends to work
The impedance of the microphone is 25 ohms and so a separate
matching transformer is necessary in 300 ohm circuits, and is
especially important if the 4017 is to be " mixed " with other types.
The output level is approximately 4 dB lower than that of the 4038
ribbon microphone. The weight is 24 lb. This microphone is
now regarded as obsolete, the replacement types being the S.T. &
C. 4032 and 4035, which weigh about 8 lb.
(6) S. [email protected] C. 402 I A and 4021F (" Apple and Biscuit ")
This microphone was designed to give omni-directional characteristics at high as well as low frequencies. The case is therefore
spherical, and reduced in size. The " biscuit " does much
towards maintaining all-round response; high fiequcncy sounds
from the top are attenuated and those from underneath are partially
reflected on to the diaphragm. This counteracts the high frequency rise due to the obstacle effect mentioned earlier.
There are few studio applications where an omni-directional
microphone is used at present, since it is more difficult to achieve a
satisfactory ratio of direct to indirect sound under reverberant
acoustic conditions than with a figure-of-eight or cardioid microphone. The 4021 finds its use in outdoor applications, in echo
rooms and as the talk-back microphone in studios.
This is designed to repl& the 4017 for hand use only, is smaller
and lighter, and is mounted in a streamlined hand-grip. I t has an
improved frequency response, but has the same general directional
properties. Its internal construction is similar to that of the 4021.
It weighs 2 lb. A windshield is available which will fit either the
4032 or the 4035.
The 4032D is identical with the above, except that a transformer
fitted inside the handle provides an output impedance of 7,000
ohms for direct connection to the E.M.I. Midget Recorder.
( d ) S.T. O" C.40354
This is the general purpose replacement for the obsolete 4017C
microphone. It, too, is smaller and lighter and has the same
directional properties with an improved frequency response. It is
useful for outdoor work of all kinds, and weighs I lb including the
cable-connecting jack.
(e) S. T. 6' C. 40374
The diaphragm and magnet system of this microphone is similar
to the 4021 or " apple and biscuit
It is fitted in a " stick "
tubular mounting about I in. in d i e t e r and 8 in. long. The
frequency response may suffer if the vent holes at the base of the
microphone are covered, although ordinary handling will not
usually cause trouble.
Two types of windshield are provided, a small plastic one made
by Standard Telephones and Cables Limited and a larger one,
BBC designed, of wire mesh. It has been found that further improvement under windy conditions is obtained if a " bung" is fitted
Plate 5. I . S.T.3 C . moiing roil [email protected]
4037, 4021, 4035 and 4032.
Ribbon microphones-lype
Below: PI& 5.4. S.T. €3 C. dypc
grog-lip ribbon microphone
.4booe and leJ: Plate 5.7. Neumann electrostatic microphone Qpe K M g g a and close-speaking
Plate 5.8. C. rz microphone, filar diagram selector and supply unit
Plate 5.9. A.K.G.C. 28 microphone and mains unit and A.K.G.C. 29, C. 30 extension pieces
Plate 5.ro. K. M. 56 microphone
inside the microphone to restrict the passage of wind through the
vent holes. The slight bass loss which is introduced by the bung
is not normally serious for speech.
The 4037A, and more particularly the shorter but otherwise
similar 4037B is inconspicuous in use, and these microphones therefore have recommended themselves to television interviewers and
dance band vocalists. The 4037A weighs approximately 2 lb with
cablejack, and the 4037B just over 4 Ib.
(f)R.C.A. BK6B
This is a small moving-coil microphone for " personal " use, and
is intended to be hung round the neck of the user, a lanyard being
provided for this purpose. The micrgphone can be conveniently
positioned beneath a man's tie, and is then reasonably inconspicuous
for television use. The frequency characteristic%of the &crophone
have been designed to compensate for the somewhat unusual position
of the microphone in relation to the mouth of the speaker and for
the fact that it may be covered by clothing.
With all personal microphones, some such correction is generally
necessary. Moreover, if the microphone has to be used in a programme where the speaker is using other static microphones, special
equalisation, depending on the studio conditions and the person
concerned, may be desirable in order to prevent too great a change in
speech quality when the two types are used consecutively.
There are seven versions of the ribbon microphone in current
use-the older types AXB and AXBT, the P.G.S., S.T. & C. Type
4038, Reslo VRM/T and finally the BBC Lip Ribbons LI and
L2 (Plates 5.2, 3 and 4).
(a) BBC-Manoni T
* AXB and AXBT
The AXB microphone, which is a BBC design, came into
service in 1934, and the later version, AXBT, in 1943. These are
identical in appearance, except for a white " T " painted on the
AXBT. The latter has a stronger magnet, and in consequence has
6 dB greater output, and musical " attack " appears to be better
reproduced. The AXB type is now being withdrawn.
In the vertical plane, the acceptance angle is much narrower for
high frequencies than for low frequencies and is not symmetrical,
so that different quality is obtained above and below the axis. The
angle of tilt is therefore very important, and angles in excess of
about 20' to the source result in a loss of top.
For discussion programmes and for use in theatres etc., the type
AXBT microphone is often felt to be too large and conspicuous.
(b) P.G.S. (Presmegadient Singlc) *
New magnetic materials have made possible the design of smaller
and lighter microphones, and the BBC Research Department
has produced the P.G.S. ribbon microphone, which is less than
one-third the weight of the AXBT (2-5 lb, compared with 9.2 lb).
Its output level is 4 dB less than the AXBT, but it has a much
improved frequency response--the bass is better maintained, and
the top extends about half an octave higher. To keep the sue
down to a minimum, the case fits closely around the magnet and the
matching transformer is inside the base of the stirrup. The ribbon
is I in. long, compared with 2) in. in the AXBT.
The variation in high frequency response with angle of tilt,
noticed in the Type AXBT microphone, is less in the P.G.S. The
production models of the P.G.S. are manufactured by Standard
Telephones and Cables Limited. Three type numbers are in
existence but only the last named will normally be met in BBC
sound studios:4038A-Black case, 30 ohms output impedance
4038ILBronze case, 30 ,,
case, 300 ,,
In BBC television studios the 4038B is in regular use since many
of the microphone circuits are of 30 ohms impedance, as opposed to
300 ohms in sound service. To help identify the microphones,
even in badly-lit conditions, the 4038B has a small recess engraved
" 30 ohms ", and the 4038C has a small raised button engraved
" 300 ohms
This is a miniature pressure-gradient ribbon microphone having
a nominal figure-of-eight response. It is very small in size and
therefore has become popular for use in television studios, where
"Single" refem to the fact that this microphone has a single hoDeahoe
magnet. The turn was used to differentiate between this microphone and an
experimentd type known as the P.G.D. (D for double) which had two magnets.
The P.G.D.was nor produced commerciallyY
it can be used as an inconspicuous " inshot " microphone. It is a
modified version of the outwardly similar Redo RBM/T.
( d ) Noise-cancelling LipRibbon [email protected] ( LI- I 937 design; L2'951 design)
For sports commentaries etc., a very close speaking distance is
necessary to discriminate against background noise. In the BBCMarconi lip-ribbon microphone LI, a guard ring gives a speaking distance of 23 in., and the resultant rise in bass response is
corrected by means of acoustic impedances (supplemented by
electrical filters as described below). This bass cut attenuates a
large part of the crowd noise and gives better speech-to-noise
In the LI, a cradle of eight springs gives protection against
bumps, and the base of the horseshoe magnet protects the ribbon
from the direct draughts from the speaker's mouth. The carryingcase is specially made to clamp the springs securely, and the
microphone should always be transported in this way.
A three-position switch in the carrying-case gives conditions as
follows :Position No. 3-unequalised, Positions 2 and ~--different degrees
of bass cut; these positions are intended for loud, average, and
quiet speech. Quiet speech requires the most bass cut, since
the frequency content of speech under these conditions is bassy
anywayAn alternative noise-cancelling microphone known as the L 2
is now in service. It is of BBC design and manufacture, and
has an improved frequency response, and 5 dB greater output. It
weighs I lb, compared with 1.8 lb for the earlier model. The
hand-grip is oval in cross-section, and the angle it makes with the
microphone may be varied to suit the individual commentator.
The cradle of springs was found to be unnecessary in this model,
provided reasonable care is taken in handling the live microphone
Production models of this lip microphone are manufactured by
Standard Telephones and Cables Limited and the type number is
41oqAA new version of this microphone has now been produced by
S.T. & C. It is designated 41oqB and has a slightly wider frequency response than the earlier models. It does not have a
separate bass equaliser, equalisation similar to the " average "
position being built into the microphone. The 41oqB can be
recognisbd by the h c t that it has only a single bar as a mouth guard.
A third version, 4104C, is similar but of 300 ohms impedance
S.T. & C @ (Oardioid) (Plate 5.6)
This microphone combines a ribbon element and a small movingcoil element of the 4021 type. A screwdriver-operated switch
gives the following conditions :-
using moving-coil only.
using ribbon only.
combining both elements.
As was stated earlier, constructional difficulties are likely when
ribbon and moving-coil units are combined in this way. Practical
mechanical difficulties prevent the moving-coil unit h m being
omni-directionala t all frequencies. A compromise is necessary, and
in this case the outputs of the moving-coil and ribbon units are
combined in a suitable electrical network in such a way that as the
frequency rises the effect of the ribbon unit is attenuated. By this
means the moving-coil element contributes the whole of the output
a t high frequencies and a good compromise cardioid response is
The rugged nature of this microphone makes it especially suitable
as a boom microphone in television. The ribbon is of unusual
stiffness to reduce mechanical bumps, and a shockproof mounting
is fitted. A matching transformer is required in 300 ohm circuits,
and gives an output level similar to the 4038. In the cardioid
position there is a small increase in bass, as would be expected with
close working, but the overall frequency characteristic is such that
the increase is unimportant at distances over g in.
All electrostatic microphones need an amplifier close to the
diaphragm. In some designs it has been possible to move it a
short distance away, up to 3 A, by the use of special very low
capacity cable. The need for an amplifier complicates and enlarges
the microphone, involves high and low voltage supplies, and a
multi-core connecting cable. In recent types miniature valves and
components have made possible reduced size, and mains units
have replaced the batteries. The types of microphone which will
be described are the Neumann KM54a and KM56, and the A.K.G.
C.rn, C.28, C.29 and C.30.
KM54a (Plate 5.7)
The K M 5 p consists of a capsule having a metal diaphragm
mounted at the end of a small cylindrical case containing the head
amplifier. The axis of the microphone is end-on, and the directional properties are nominally cardioid, but the high frequency
response is found to change somewhat at oblique angles. The
rejection of unwanted sounds from the back is a very desirable
feature in multi-microphone layouts.
A windshield may be necessary, and two types are available. The
makers' windshield, described by them as a " close talking shield ",
modifies the response and polar diagram, introducing some bass cut,
which may be an advantage in the case of " crooners
alternative type, a BBC design, does not impair the frequency
response, and can safely be used when close working is not required,
without affecting the quality. When fitting either windshield care
should be taken to see that it is pushed gently on to the microphone as far as it will go.
(b) A.K.G. C. I 2
This microphone is a product of the Akustische und Kinogerate
G.m.b.H. of Vienna. It is an electrostatic microphone carrying
effectively two cardioid capsules with a common back plate, this
combined capsule being spring mounted in a long narrow case
containing a head amplifier. The microphone axis is side-on to
the amplifier case.
The head ampUer/microphone unit is connected by 65 ft of
cable to a mains unit supplying polarising voltage for the microphone capsule and operating voltages for the amplifier. A further
cable leads h m this mains unit to a polar diagram selector unit
and the output of the microphone is taken from a socket on this
unit (Plate 5.8).
The polar diagram selection unit allows remote selection of the
directional pattern of the microphone from omni-directional
through cardioid to figuresf-eight, with three intermediate conditions between each-nine switch positions in all. The operation of
this polar diagram selector will be described with reference to Fig.
5.13. It will be seen that the fixed back plate of the two capsules
is maintained at a steady potential, exactly half of that of the
available polarising voltage. If the available voltage is IOO volts
this fixed plate is thus maintained a t 50 volts. The fiont diaphragm of the microphone is at zero potential, that is, it is earthed,
and so a constant 50 volts difference is maintained on the fiont half
of the microphone. The rear diaphragm of the microphone is
connected to the slider of a linear potentiometer which is connected across the whole of the available polarising voltage so that
this diaphragm can have any voltage between zero and one
hundred applied to it.
It will thus be seen that the diaphragm
can be either 50 volts positive, zero, or 50 volts negative, with
respect to the centre plate. Obviously, intermediate positions are
also possible.
(c) A.K.G. C.28 (Plate 5.9)
This microphone is a single diaphragm electrostatic microphone
similar in operation to the KM54. The rnicrophone/head amplifier
unit is rather larger than the KM54, but not so large as the C.12.
The polar response is cardioid, with the microphone axis " end on "
to the case.
The microphone has a good bass response and a smooth high
frequency response. A windshield, W.28, is available, but even
without it the C.28 is not seriously prone to " popping " in close use.
The actual voltage is 105,but mund figurca have been wed to simplify the
(d) A.K.G. C.26
This is similar in appearance to the C.28, but has a capsule which
is omni-directional in characteristic.
A.K.G. C.29 and C.30 (Plate 5.9)
These are similar in performance to the C.28, but an extension piece has been fitted between the microphone capsule and the
head amplifier unit. In the C.ng this extension is 12 in. long,
and in the C.30 it is 36 in. long. These microphones are useful
as stage microphones where an unobstrusive instrument is desirable
so as not to block the vision of an artist from the audience. A windshield is available for both.
A possible source of confusion arises in that microphones of the
outward appearance of the C.29 and C.30 can in fact accommodate
the omni-directional capsule normally fitted to the C.26; in fact
all the parts of these microphones are physically interchangeable.
The different types can be recognised with certainty only by observing the number engraved on the capsule after removing the windshield.
(f)W . n KM56 (Plate 5. I 0)
This is similar in appearance to the KM54a microphone, but has
a double cardioid capsule similar to that of the C.1 2, and the axis
of the microphone is " side-on
This enables it to have variable
polar diagram characteristics, and the control for the polar diagram
is found on the microphone itself. Unlike the C. 12 no intermediate
switch positions are available.
Crystal Microphones
Crystal microphones are not met a great deal in broadcasting
work, but are often found with domestic tape recorders. Some
types, however, are used by the BBC, and these will be described.
This microphone is used in Mobile and O.B. work, where a
commentator must be able to move about and have both hands free.
It has a rubber-covered case about 14 in. square, which pins on
to the clothing. There is good insulation against mechanical
shock, but the cable itself should be pinned to the wearer's clothing
a few inches from the microphone to prevent it generating noise.
A short length of cable connects the microphone to the pocket preamplifier. The power supplies for this pre-amplifier are provided
by a 15 volt battery, and a single pen cell. There is an ON/OFF
switch, and a total of 5 hours' continuous operation is possible. As
the batteriu are cheap, it is recommended that a fresh set be fined
%r each programme commitment, d i s c a d d batteries being used
up on rehearsals.
(b) Acos Mic.gg
This is a small " stick " type microphone, with the crystal
capsule mounted a t the end of a plastic tubular case about 5 in.
long. It is used in the BBC with the Self-operated Outside
Broadcast Equipment.
(c) Ronetk MC65
This is a crystal microphone capsule which for BBC use has been
mounted in a plastic case of the " stick " type about 5 in. long.
The microphone is of high impedance and has been used by the
BBC with miniature tape recorders.
( d ) Double Lapel Microphone MCI4
This consists of two Ronette crystal capsules each mounted in a
small metal case with a clip, intended to be fixed to the lapels of a
man's jacket, one on each side. The capsules are connected in
parallel, and this arrangement helps to prevent loss of volume as
the speaker's head is turned from side to side. The output impedance is again high, and the use of the miniature pocket pre-amplifier
converts sound energy into electrical energy, and
this outpyt " follows " the sound in magnitude and rate of vibration.
A loudspeaker reverses this process, being driven by the electrical
energy (suitably amplified), and made to vibrate in the same
manner as the original source of sound.
The design of a loudspeaker to radiate a wide range of fre-,
quencies presents a number of technical problems, and some of these
are briefly discussed, followed by short descriptions of BBC
loudspeakers in current use.
The moving-coil loudspeaker employs a strong magnet, specially
shaped to concentrate the magnetic field in a narrow, ring-shaped
Fig. 6.1.
gap. The speech coil is suspended in this gap and fastened to a
conical diaphragm which is usually moulded from paper pulp (Fig.
When programme currents are passed through the coil, interaction with the field of the magnet sets the mil and cone into
vibration. The cone will then radiate sound waves a t the frequencies present in the current.
How closely this radiation reproduces the original sound is, of
course, a matter of careful design.
The h t complication in achieving the ideal loudspeaker is with
regard to the directional pattern. As was seen in Chapter I, a
radiator of waves will propagate in all directions or in a relatively
narrow beam, according to whether it is small or large compared
with the radiated wavelength. But a normal loudspeaker is intermediate in size in relation to the range of sound wavelengths (for
50 CISand IO,OOO cis the wavelength in air is about 22 ft. and
14 in. respectively). This means that low frequency sounds are
sent out in all directions, while higher frequencies tend to be confined to a narrow angle about the axis of the loudspeaker (Fig. 6.2).
The effect on listening is that best results are obtained directly
in front of the loudspeaker, and listening at the sidk gives the effect
of " top cut
Furthermore, reverberant sound due to the
existing room acoustic will have very little " top
The M e 0s Enchure
Another factor which limits the efficiency of a loudspeaker, particularly at low frequencies, is the interference between the back and
front radiations. Referring to Fig. 6.1 we see that a forward movement of the cone will cause a compression of the air in front, and
an expansion at the back. The net result is the simultaneous
radiation of two waves in anti-phase, which will cancel each other
unless the path from back to h n t is made greater than half-awavelength for the lowest frequency required.
A 2 fi 6 in.-square ba& (see LSU/4A, Section 6.3.1.) will
prevent this cancellation for frequencies down to about 200 c/s
(wavelength 5 ft), but short of building the loudspeaker into a wall,
a b d e provides only a limited solution.
The addition of sides to form a box b a f i can give improved bass
efficiency, provided steps are taken to reduce the boominess due
to resonance of the enclosed air-for example, by the insertion of a
lining of glass wool. The back of such baffles is left open to reduce
resonance. A vented enclosure, however (see LSU/IO, Section 6.3.2.)
has a closed back, and an aperture by means of which the low
frequency resonances can be controlled (Fig. 6.3).
For efficient radiation of low frequencies the cone should be
large, but the time taken for the coil vibrations to travel out to the
edge of a large cone may correspond to one or more cycles at high
frequencies. This means that parts of the cone are radiating in
anti-phase, with consequent loss of efficiency.
I t is for this reason that all wide range monitoring loudspeake.g. LSU/I~-incorporate an auxiliary high frequency unit. An
electrical filter splits the output of the amplifier, sending low frequencies to the large cone or " woofer ",and high frequencies to
Fig. 6.3. Lma3prakrr aulanau: (a) b+;
(6) box cabinet; (c) wnkd cabinet
the " tweeter ",which may be of the moving-coil type or work on
the electrostatic principle described in section 6.2 (Fig. 64).
Just as there are moving-coil microphones and moving-coil loudspeakers, so the electrostatic microphone has its countapart in the
electrostatic loudspeaker. This principle has been known for
many years, and has been used for tweeter units, but until comparatively recently no attempt has been made to produce an electrostatic loudspeaker covering the full audio frequency range. The
early electrostatic loudspeakers consisted of a fixed back plate and
a conducting diaphragm to which a high polarising voltage was
applied, thus creating a charge on the capacitor. The programme
voltages were also applied to the capacitor, and the diaphragm
would therefore move, reproducing the sound (Fig. 6.5). Unfortunately, as the diaphragm moved in relation to the back plate its
distance from it varied, and since the charge on a capacitor supplied
at constant voltage depends on the distance between the plates, this
varied also. Hence the driving force on the diaphragm varied with
amplitude of the signal and so distortion was inevitable. A further
difficulty arose in that it was not possible to make the movement
of the diaphragm sufficiently free at the same time as preventing it
from collapsing on the back plate under the influence of the pull
due to the polarising voltage, destroying the capacitor. Such
loudspeakers were only possible for use at high frequencieswhere the
amplitude of movement of the diaphragm was sufficiently small to
Fig. 6.4. Us of "kuccta
make the distortion negligible. The small diaphragm movement
made it easier to construct a unit that would not collapse.
The recent development of wide range electrostatic loudspeakers
has come because of the invention of the " constant charge"
principle (Fig. 6.6). In this case the polarising voltage is applied
via a high resistance which tends to prevent fluctuations in charge
as the diaphragm moves. The situation can be further eased by
mounting the diaphragm between two fixed plates so that the two
forces of attraction tend to cancel out. The fixed plates must be
perforated in order to allow passage of the sound waves.
As the diaphragm flexes, the gap, and hence the capacitance
between it and the plate, will vary, more in the centre than at the
outside. It is therefore necessary to make the conducting coating
on the diaphragm and b e d plates of a high resistance material to
prevent the charge fiom redistributing itseras the cliaphragmmoves.
These improvements made the use of bigger diaphragms possible,
and loudspeakers of this type are claimed to have wide &equency
response, and extremely low distortion. A typicd version has a
figUtesfeight response at low frequencies and this may be beneficial
in rooms with poor acoustics since the eigentones will not be excited
on the dead sides of the loudspeaker. (See Figure-of-eight Microphones, Chapter 5.) There are disadvantages in that a special
unit is required to supply the polarising voltage which may be
several thousand volts, and a transformer is necessary to match the
output of an amplifier, normally of about 15 ohms impedance, to
the extremely high impedance of the loudspeaker.
A further disadvantage is that the maximum sound output for a
given size of loudspeaker is somewhat less than that obtained fiom
moving-coil loudspeakers of comparable dimensions, and the overall efficiency is lower, due to the difficulty of matching the unit to
the amplifier; the electrostatic unit needs more power for the same
loudness than the moving-coil one.
The following loudspeaker units are in current BBC use and
since their designs and applicationsare similar to those used by other
organisations they will be described as suitable examples of each
This is a baffletype loudspeaker of medium quality reproduction
used mainly for talk-back. A 10 in. loudspeaker unit is mounted
on a 2 ft 6 in.-square baffle and its associated amplifier is
mounted on the baffle with its volume control coming through the
fiont. The whole assembly can either be stood on the floor or
mounted on a wall bracket. The input impedance of the amplifier
is ~o,oooohms and so can bridge a 600 ohm circuit without affecting
its level.
Because of the relatively small size of the baffle the bass response is
rather poor and so this loudspeaker is normally used for speech.
This is the standard BBC high quality monitoring loudspeaker
(see Plate 6. I).
The loudspeaker has two concentric units, and a third unit which
radiates extremely high frequencies. The large 15 in. cone is used
for frequencies below 1,200 CIS. The second unit has a 14 in.
diaphragm which radiates via a tapered hole in the magnet
pole-piece, through a honeycombed horn. This deals with fiequencies up to about 7,000 c/s. The high frequency unit, which is
a separate moving-coil diaphragm unit, with a plastic cone, carries
on above this and takes the axial response to well over 15,000 CIS.
The cabinet, developed in BBC Research Department, is of
the vented type described earlier. Three laym of carpet-felt
stretched horizontally across the centre of the cabinet suppress
up-and-down resonances, and together with lagging-of the enclosure
prevent high frequency radiations from the vent.
The amplifier is a commercial product, and is housed in a
shelved comDartment at the right-hand side of the cabinet. The
input impedince is sufficiently g g h to permit operation on 600 ohm
circuits without loss of level, and the maximum power output, for
an input of - 20 dB, is 10 watts.
The height of the open vent above floor-level affects the bass
response, and should not be altered-for example, the fitting of
castors is not advisable. These were deliberatelv left out of the
original design for reasons of safety, it being felt that such a heavy
piece of hrniture should be moved only under supervision.
6.3.3. I&j/x (Plate 6.2)
This loudspeaker, ina totally enclosed cabinet considerablysmaller
than the LSU/IO, was originally designed for high quality monitoring on Outside Broadcasts. I t is, however, being increasingly used
as a studio monitoring loudspeaker, since its small size makes it
easier to accommodate than the LSU/xo.
The loudspeaker is used with a separate power amplifier AM8/1,
which has a similar input impedance to that of the LSU/xo, so that
again this can bridge a 600 ohm circuit without affecting level.
The axial Erequency response is excellent, extending smoothly to
1 3 , m CIS, but, due to the relatively small size of the cabinet there
is a slight, but not serious, bass loss.
The loudspeaker contains three units--a
15 in. bass unit and
two identical high frequency units.
6.3.4. ISdx (Plate 6.3)
This loudspeaker is the development from the prototype known
as the LSU/IIA, described in a paper presented at the Institution of
Electrical Engineers in 1958.
I t is a studio monitoring loudspeaker of rather larger dimensions
than the LSQII,and has improved frequency characteristics. The
power amplifier, AM8/4, is mounted under the loudspeaker cabinet
which is of the vented type. The loudspeaker units are the same
types as those in the LSQII loudspeaker; a 15 in. bass unit and
two high 6-equency units.
Electrostatic loudspeakers have been used as monitoring speakers
in studio cubicles but are not in general use. It has b u n found
that the maximum sound level produced by existing units has not
always been sufficient for studio requirements; further difficulties
arise in that the loudspeakers cannot be placed close to a wall,
due to their figure-of-eight radiation pattern, and have to be mounted
on the floor if loss of bass is not to occur.
The relative sensitivity of the human ear to sounds at different
frequencies changes when the level of listening is changed. If, for
example, the volume control of a loudspeaker is turned up, the
extreme bass and top will be heard to increase more than the
middle. Thus, the " balance " of the different frequencies has been
upset, in other words we have introduced attmuation dhtmtion (see
Chapter 12). It is, of course, equally true that listening at lower
levels will reduce the high and low frequencies more than the
middle and this is often the case under domestic conditions during
periods of background listening.
Since the balance of frequencies, as heard by the ear, depends on
the listening level, it is important that the studio manager and producer should monitor programmes a t the correct volume. How
are they to decide what this volume is? Let us look at the different
types of programme and see how this decision can be made. In
serious music, for example, in the case of a symphony orchestra,
the level at the studio manager's ear in the listening cubicle should
be as nearly as possible the same sound level as he would hear
standing near the main microphone in the studio. This is only
strictly true so long as the number of microphones used is few, but
nevertheless in multi-microphone balances in this type of music
this level will give good results. In the case of chamber music it is
seldom that more than one microphone need be employed, so
again the sound level at that microphone is the one to choose. It
must be borne in mind that many listeners will not be listening at
such a high level as this, and checks should be made at a lower level
to ensure that the balance remains satisfactory.
Other types of music, however, are not so simple. In the case
of dance bands and modern light music, convention and modem
musical arrangers decree a balance which may have anything up to
2 0 microphones, none of which strictly speaking, can be termed the
main microphone, and the sound balance in the studio is not
Plate 6.1. Studio control cubicle showing loudspeaker unit LSUIro
Plate 6.2. C o r w of s f d w cubicIe showing loudcpakm unit =3/1
Plate 6.3. Loudspeaker unil
necessarily the balance of sound heard in the listening cubicle. In
this case the studio manager and producer should set the level of
their loudspeaker to give a suitable volume considering the size
of the instrumental combination and the type of music being played;
again bearing in mind that the average listener may well be listening
at a much lower level than this and may therefore lose the effect
of some of the high and low frequency components.
A step by step volume control switch can be fitted to the studio
desk and recommended settings in decibels are: - 6, -4, -2,
normal, +2, +q, +6, +8, +IO.
The " normal" setting should
be adjusted on the loudspeaker control depending on the particular
type of programme. The advantage of the switch is the ease with
which the volume can be returned to a previous setting when it has
been changed for the purpose of checking at a level more nearly
that at which an average listener will hear the programme.
Correct Use of the Loudspeaker
The following points should be borne in mind when using a
monitoring loudspeaker:
(a) The listening level should be as nearly as possible that which
could be heard in the studio as described above.
(b) Frequent checks should be made during rehearsal at a level
more nearly that at which the average listener would hear the
programme; usually somewhat lower than the level defined in (a).
(c) In the mixing of sound effects and in deciding on the levels
for small sounding instruments, such as the clavichord, allowances
should be made for the additional background noise on the
listener's set.
(d) Monitoring of quality should be carried out near the axis of the
loudspeaker, because of the directional effect of high frequencies.
(e) Listening distance should be not less than 3 ft but not
greater than 6 ft, since the loudspeaker quality and the
apparent reverberation may be seriously modified by the acoustic
properties of the listening room. Furthermore, if the loudspeaker has separate high and low frequency units, of necessity
some small distance apart in the cabinet, listening too close to
the loudspeaker will not give a correctly integrated sound. Loudspeakers with coaxially mounted units do, of course, not suffer
from this difficulty.
(f) There should be no obstruction between the studio manager
and the loudspeaker.
THEStudio Control Desk provides facilities for mixing a number
of sound sources, e.g. microphones, gramophone turn-tables, tapereproducers and outside sources; controlling their overall volume;
monitoring the studio output, and cueing the artists. In this
chapter the requirements of such a desk will be described and
some examples given from BBC and other equipment.
It is extremely important in this chapter to realize that certain
diagrams have been simplified to illustrate separate points. The
symbols used to illustrate some components are shown in Fig. 7.1.
The requirements of any studio controlling equipment can be
summarised under seven headings as follows:I.
Means of mixing various sources:(a)
Microphone sources.
Gramophone pick-up sources.
Tape reproducers.
Outside sources from other studios or organisations.
Group and channel switching and control of overall volume.
3. Addition of artificial echo or reverberation.
4. Measurement of programme volume.
5. Means of listening to the studio output.
6. Talk-back system into the studio, and possibly to outside sources.
7. A cue system to the studio, preferably by means of lights.
Each of these requirements will be described in detail.
7.1.1, M~SSM
of Mixing Various Sources: Faders
In order to enable the various sources to be mixed in the differing
proportions needed to achieve a satisfactory sound balance, each
source must be provided with a fader so that its level can be adjusted.
Faders are variable attenuators and can take several forms. The
simplest form is a variable resistor either in one leg or both legs of
the microphone circuit, and until fairly recently many such faders
were in use in broadcasting organisations. Fig. 7.2 shows a form
Fig. 7.1. Circuit p t b o l s
of fader often used in domestic and semi-professional equipmentthe shunt potentiometer which has many applications. Figs. 7.2
and 7.3 both show unbalanced attenuators, i.e. the resistance is only
in one leg of the programme line. A balamed circuit is one in which
both legs are symmetrically disposed above earth. This form of
circuit is often used for carrying programmes-especially at low
volume-in order that unwanted currents, such as hum, induced in
the two halves by stray electric fields etc. will be identical. Since
the currents will be in opposite sense in the closed circuit, they will
then cancel. A circuit of this type is shown in Fig. 7.4 with a
balanced potentiometer having two sets of sliding contacts ganged
together. Both of these simple types of fader, whilst perfectly
satisfactory in themselves, can cause difficulties when several are
wired together as a mixer (Fig. 7.5). The main difficulty arises
because as the fader is moved from minimum to maximum the
impedance seen by the following equipment changes considerably,
and when the outputs of a number of such faders are combined
Ri&: Fig. 7.2.
Unboloncsd saiar attawator
7.3. Unbalmdd
hunt ntunuatw
Fig. 7.4. Exmfilc of a balanctd circuil
Fig. 7.5. Simp& m i w wing saissfadas
Fig. 7.7. Plug and jack show
ulg ''UIMI" EOnt(DCt(
this impedance change becomes sufficiently magnified to cause a
change in overall volume which may make mixing very difficult.
The effect, however, can be minimised by never fading the channels
up fully, that is keeping some resistance in circuit to act as a buffer
between the channels.
A more satisfactory type of fader is the constant-impedance
fader which, as its name implies, presents the same impedance to
the following equipment no matter in what position it may be set
(Fig. 7.6). Such faders are commonly used in professional equipment, but in view of their complex design are normally too costly
for domestic use.
Croup and cbanud Switching
It is obviously desirable, particularly in a permanent studio
installation where microphone points are Gxed to a wall skirting,
that means should be provided for any given microphone point
to be plugged or switched to any channel on the desk. For this
purpose plugs and jacks are commonly used and by their means
flexible connections in the momamme
circuits can easilv be made.
In Fig. 7.7, the plug shown has three connecting points, the tip,
ring and sleeve. The programme circuits are connected to the tip
and ring and the earthed screening to the sleeve. In the jack
inner contacts are sometimes wired so that a complete circuit
exists. Inserting a plug breaks this circuit, and the plug circuit
then replaces that of the inner contact. This type of" break jack "
is used for cross-plugging amplifiers, etc. Some trouble has been
experienced at extremely low microphone volumes with circuits
made in this way, and it has been found more satisfactory always
to plug circuits at this level.
Having arranged the switching of sources to channel faders, it
may be desirable to group a number of these channels together to
a hrther fader, so as to be able to adjust them with one control, the
I 08
remainder of the channels on the desk either being left as individual
channels or connected to another group control. One of these
individual channels can then be used for a narrator independently
of the groups. The outputs of the groups must then be fed to a final
main gain control, which is used to set the overall volume of the
programme. Switching of channels to groups can again be done
by plugs and jacks, or by conventional switches (Fig. 7.8). The
main control, since it will be used during transmission or recording
to vary the overall volume of the programme, should be graduated
in small steps so that no abrupt changes occur as the control is
moved. In practice it has been found that with n o d programmes,
steps of 2 dB are generally considered to be satisfactory. Of course,
if the fader is continuously variable as opposed to a stud type, no
difficulty arises.
7.1.3. Artificial Reverberation (Echo)
In studios for music, particularly light music and dance music,
and in studios used for dramatic productions, a means of adding
artificial reverberation to any given channel is desirable. Control
should be provided over the proportion of direct to reverberant
Fig. 7.9. Reuerberotion room,
showing mult;Ple wall refictwns
sound on any given channel and an overall control on the output
of the reverberation device. The system works as follows:
A hybrid transformer, which is a transformer having one input
and two identical and independent outputs, is connected after the
1 0 ECHO
Fig. 7.10. Sirnplijkd schematic of magnetic d m r e ( ~ b e r a t hdsvi4c
channel fader. The two outputs of the hybrid go to the two halves
of a ganged rotary " mixture " switch which controls their levels
in opposite senses. The hybrid transformer is necessary so that
the output of the reverberation device does not feed back via the
" direct " side of the mixture switch to the input, thereby causing a
howl-round. One output goes to the group fader in the normal
way, and the other to the reverberation device. The output of the
reverberation device is fed back to the group via a reverberation
Reverberation equipment can be of two types, a room or an
electronic device. A reverberation room is simply a bare room
having highly reflective walls, in which are placed a loudspeaker
and a microphone, the layout being designed so that little direct
sound from the loudspeaker can reach the microphone (see Fig.
7.9). The signal from the hybrid transformer and mixture switch
is fed to the loudspeaker, and the reverberation produced in the
room is fed back via the microphone. Various types of electronic
reverberation machines exist. One of these depends upon a magnetic recording of the original sound being played past a number of
replay heads, and the output of these heads being fed back into the
chain. By this means a number of discrete echoes are produced and
by feeding the output of the last of these heads back to-the recording
head, these echoes can go on, theoretically a t least, to infinity
(see Fig. 7.10). This device is not too satisfactory since it is often
possible to hear individual echoes, particularly when long reverberation times are desired. Another type of electronic reverberation
I I0
machine consists of a large rectangular thin steel plate similar in
some respects to the traditional theatre thunder sheet. This plate is
mounted vertically, tensioned at its four corners. A moving-tail
drive unit, similar to that used in a loudspeaker, mounted near one
end, causes the plate to vibrate, and a contact microphone, mounted
near the other end, picks up these vibrations some time later.
The output of this microphone is fed through a suitable equalising
Pig. 7.rr. Plate 3rps rwarberation dmicr
amplifier back to the programme chain as shown in Fig. 7. I I. The
equalising amplifier is necessary in order to remove the effects of
the various mechanical resonances of the system. Control over
reverberation time is possible by moving acoustic damping material
nearer to, or farther from, the plate surface. A disadvantage of
the plate system is that at the longer reverberation times the bass
response tends to rise. This is not, however, serious at the reverberation times most normally used.
7.14. Marcmrement of Propamme Volume
I n order that the average volume leaving any studio shall be the
same as that from another sound source making up a given programme, it is necessary that some form of measurement of the
output levels should be made. This measurement wilI also give an
indication when the safety limits of the system are likely to be
exceeded either at the high end of the scale, where the onset of
distortion must be prevented, or the low end where small signals
can cause a poor signal-to-noise ratio. Various devices have been
proposed for measuring the programme and two will be described.
The first of these,. the volume indicator or m l u m unit meter, shown
in Plate 7.1(a),' is used by many organisations particularly in
America. I t consists of a rectifier-moving-coil meter having
specially designed ballistics. On programme, its readings are
arbitrary and the calibration shows " percentage utilisation of
The scale is calibrated 0-100 with the 100 mark at
the channel
about two-thirds full scale, the portion of the scale above this being
coloured red. The second scale, calibrated in decibels, is for use only
on steady tone when the ballistics of the meter can be ignored. This
meter is reputed to measure average programme volume, but in
fact its readings are purely arbitrary, and it can give misleading
results. The reading depends on the type of programme material;
a sustained loud programme may well give a lower reading than
quieter impulsive sounds. Furthermore, some difficulty is enwuntered in reading the instrument since the " fall time " of the meter
needle is fairly fast, the resulting rapid movement up and down
being rather diffcult to read. The BBC, therefore, uses rather
a different meter, the peak programme meter (Plate 7.1(b) ). In this
instrument, which is a form of valve voltmeter, the circuit is
designed so that the meter needle will rise quickly to the peak
value of the programme, or very nearly so, and then will fall again
slowly. By this means the rapid movement of the needle to and
fro, as in the volume unit meter, is avoided and a more even and
easily seen indication results. The rise time of the instrument is
approximately -004 seconds and the die-away time approximately
3 seconds. In order to reduce eye strain the scale has been printed
white on black and the number of divisions reduced to seven, each
division corresponding to a 4 dB change in level.
In BBC practice lineup tone at zero programme level should
give a reading of" 4 " on the P.P.M. and this is normally arranged
to correspond with a 40 per cent modulation of the transmitter;
peaks of" 6 " will then result in 111 modulation. Peaks above this
can cause considerable distortion.
Mans of Ltteniug to the Studio Output
I t is obviously essential that means should be provided to enable
the overall output of the studio to be heard by the person operating
the studio desk. This can be by means of headphones, but since
the highest possible quality of reproduction is necessary, and since
a number of other people may be present, the loudspeaker is
essential. Furthermore, since the programme will eventually be
heard on a loudspeaker in a listener's home, it is desirable that
similar conditions should obtain at the time of balance, since
I I2
acoustic conditions may well affect the balance techniques. Whichever form is used it is necessary that it should not cause any loading
on the outgoing circuit of the studio, otherwise the P.P.M. levels
would be meaningless. Where the P.P.M. is used a convenient
output from the P.P.M. amplifier can be taken to the monitoring
loudspeaker. The input to this amplifier is arranged to have a high
" bridging " impedance.
7.1.6. Talk-back Facilities
A separate system of microphone and loudspeaker is necessary
between the studio control desk and the studio itself for communication during rehearsal and transmission. It is convenient for this
purpose to use the main amplifiers of the control desk, since in this
way talk-back to the studio will also be fed along the output circuit
of the studio when this is destined for a recording channel. Safeguards must of course be provided to prevent this happening if the
studio is making a live transmission, otherwise the talk-back speech
would go over the air. In this condition it should be possible to talk
to the studio only, on the headphone circuit, and on the loudspeaker
when the microphones are faded out. Talk-back to outside sources,
where these are used, may be desirable and this would normally be
carried along the control line associated with this circuit.
A number of cue lights in the studio, switchable from the control
desk, should be provided, preferably one for each microphone point.
It is convenient if these can be operated by a key adjacent to the
fader of the microphone concerned. This will necessitate switching
for cue lights in order to select key positions, since the microphone
point in the studio will already have been plugged to a particular
7.2.1. Programme Ring&
switch (HV/7)
When a studio is about to join or leave a particular programme
service, it is desirable for it to be able to hear the preceding and
following contribution. For this purpose a twelve-way ring-main
is provided in all BBC studios.
A key, normally coloured white, is provided to enable a quick
switch of the monitoring loudspeaker to be made from the ring-main
point selected to the studio output (Fig. 7.12).
I t may be necessary, when working with outside sources, to provide a clean feed of the programme leaving the studio, to be
fed back to the outside source. By " clean feed " is meant everything passing through the studio desk, except the contribution from
0 0 0
Fig. 7.12. Programme ring-main switch (HVI7). LS
Mcta s&h operates in studio when microphone L faded up
the source to which the clean feed is sent. Two reasons make
this necessary. If the contributing source is a great distance away,
say the other side of the Atlantic, it will need to hear the cue
programme from the BBC studio, say on a pair of headphones or
even a loudspeaker. If, when the BBC switches over to his
contribution, the speaker hears his own voice back, there can be a
long distance howl-round if a loudspeaker is in use, and if he is
listening on headphones his own voice will appear with an appreciable delay, due to the length of the line. This delay makes
normal speech delivery rather difficult, so under these conditions
clean feed is desirable. The second use for clean feed is when the
remote contributor is broadcasting simultaneously in his own
country both his speech and the output of the BBC studio, a
genuine two-way programme; in this case the need for prevention
of howl-round is obvious.
In studios where audiences are present, it is often necessary to
provide feeds for public address equipment h m one or more of
the microphones being used for transmission. This is done by
means of hybrid transformers in the microphone circuits and a
fig. 7.13. Switching of individual chmurcLr to Public Address
series of switches, one per channel, to enable the microphones
required to be switched to the public address (P.A.) amplifier.
Two positions of each switch give normal level and
10 dB feeds,
and an overall P.A. level control is also provided (Fig. 7-13).
Type A Studio Desk-General
The complete Type A console comprises a control desk, a studio
control cabinet, and a studio supply cabinet.
(a) The Supply Cabinet houses the mains isolators and contactbreakers.
The equipment isolator has three positions:-
I. Remote Control-when mains supply is fed via a " Mains
On ", relay operated by the red, green and orange pushbuttons on the desk.
11. Off.
111. Direct Control-when mains supply is fed direct to the
equipment; this position is to be used in the event of
failure of the go-volt relay supply.
(b) 27E Control Cabinet (see Plate 7.2) houses all amplifiers and
relays, and a pre-mixer jackfield for cross-plugging purposes.
Access to the jackfield is possible by a small door at the bottom
right-hand corner of the cabinet.
(c) 27E Control Desk has the various controls mounted on three.
upright panels as follows (see Plate 7.3) :Left-hand panel :
Amplifier change-over keys
Local red light key
Push-buttons, for ON/OFF, etc.
Standby loudspeaker key
Monitor to line key
Echo mixture switches (if required)
Echo selector key, when alternative or shared echo facilities are
Clean-feed selector key
Centre panel:
Mixing and control potentiometers
Programme meter
Cue light keys (and indicator lamps)
Echo-cut key (if required)
Right-hand panel :
Top row:
Control-room buzzer
Middle row:
Loudspeaker dim key
Telephone I ring/recall key with white indicator light
Remote record key with red indicator light
Two outside source groups, each containing:
Call button
Pre-fade listening button
Recall/answer key
Cue to control line or cue line key
Programme selector key giving pre-fade or studio output on
[email protected]
I 16
Bottom row:
Three local tape record/replay keys
Clean-feed talk-back key
Studio talk-back key
Two outside source groups if fitted
Studio loudspeaker ON/OFF switch
In addition:
Acoustic effects volume control
Cubicle loudspeaker volume control
Studio programme selector switch
Cubicle programme selector switch
Two pre-fade headphone jacks
The left-hand telephone compartment also houses the standby
loudspeaker, and the headphone jacks, which are connected to
the " studio " programme selector switch.
Two basic forms of Type A Console are in use, with the following
facilities, which can be built upon as required:Mark 11-5 channels.
Mark V-7 channels or sometimes g, echo facilities, group and
independent faders.
There is a single Mark VII console in Piccadilly Studio I, providing I 2 channels, and special arrangements for grouping of subchannels. The notes which follow refer to the Standard Mark I1
and V equipments.
Circuit Details (Fig. 7.14)
The Type A equipment was introduced shortly after the war, and
incorporated a number of novel features:(a)
Constant-impedance faders.
Individual microphone amplifiers.
Cross-plugging of sources and channels.
Push-button selection of rehearsal, line-up and transmission
(e) Comprehensive echo facilities (if required).
(f) Standby amplifiers and change-over keys.
(g) Arrangements for clean-feed working.
These will be considered in order.
(.a.) Conrtantimpedatlce Faders
All the potentiometers on the Type A desk are designed to present
the same impedance (60022) in and out. The circuit is as shown in
Fig. 7.6.
The microphone and group faders have twenty steps of 2 dBincreasing to 8 dB at the lower end. Fading out also feeds programme to the studio loudspeaker, by means of an extra contact
MIC. 1
Fig. 7.14. Simplijed diagram of TupC A Mmk I .
on the last stud. A faulty fader may be removed and another
substituted in a few seconds. But note that in certain studios,
where extra mixers have been installed, the group faders are of a
special type and cannot be interchanged with standard faders.
The Main Control fader has 30 steps, from 2 dB to 5 dB at the
lower end.
(b) Indiuiduui Microphone AmpliJies
The introduction of constant-impedance faders, as just described,
plus a special circuit of fixed resistors, ensures that accurate matching
of a given number of channels is maintained, however many of the
channels are faded up. This eliminates the drop in programme
volume which accompanies fading-in additional channels in older
equipment, but involves an inherent fixed loss of about 30 dB.
Such a loss would be very serious on programme at microphone
volume, because of noise, and pre-amplification is therefore necessary. For this reason, an amplifier is included in each microphone
circuit and the gramophone circuit, before the fader. The gain of
this " A " amplifier is 50 dB, so that mixing is carried out at about
30 dB instead of
80 dB.
Following the mixing circuits are two control amplifiers, BI and
B2, between which comes the main control potentiometer. The
output of the B2 amplifier is at zero volume, and feeds the programme line to the control room. The B2 also connects to the
programme selector switch point I I and the programme meter, via
a monitoring amplifier.
of Sources and Channels
The outputs of the A amplifiers are taken to the channel faders
via a jackfield in the control cabinet. This permits cross-plugging
(c) Cross+lugging
fig. 7.15. Action of announcer's
& in news studio
of any source-mic.,
gram., or outside source-to any fader, to
simplify operation.
An exception to this exists in the News studios. Mic. I in this
case is normally connected to the control amplifiers through an
announcer's microphone key, and not via a channel. If mixing is
necessary, Mic. I is plugged to a channel, the " censor " key then
being inoperative (Fig. 7. 15).
(d) Push-button Selection of Conditiom
Rehearsal Conditions: Pressing the green button operates the
" Mains On " relay, and provides full talk-back facilities. Operating
the talk-back key then connects the tabback microphone to the
control BI amplifier, short-circuits the main control potentiometer,
cuts the cubicle loudspeaker, and feeds the headphones and studio
Fig. 7.16. Action of talk-back kry in rehearsal caulition
loudspeaker, whether microphones are faded up or not. Note that
talk-back volume does not depend on the main control setting, and
that the studio loudspeaker programme selector must be at point
I I (Fig. 7.16). (If the selector is switched to a cue programme, this
will be heard in the studio when the talk-back key is pressed.)
Line-up Condition: Pressing the orange button operates the
" Mains On " relay, applies 1,000 c/s tone to the input of the BI
amplifier and short-circuits the main control potentiometer. The
tone has been suitably attenuated to give zero level into the programme line. This level does not depend on the setting of the main
control potentiometer, and provides a check on the gain of the
control amplifiers (Fig. 7. I 7).
Tr(uumisJion Cotulilion: Pressing the red button operates the
" Mains On " relay, and provides restricted talk-back facilities
Fig. 7.17. [email protected] condition
Fig. 7.18. Action of &-back key in transmirrion condition
Fig. 7.19. Deridon of a r t i j i d reverbnationfad in TyFe A Mark V
(Fig. 7. 1 8). Operating the talk-back key then connects the talk-back
microphone to the standby BI and B2 amplifiers, dims the cubicle
loudspeaker, and feeds the headphones and studio loudspeakervia the loudspeaker cut relay operated by the microphone faders.
Note that talk-back is at fixed volume, does not enter the programme
line and appears on the studio loudspeaker only if all microphones
are faded out. In this case, the studio loudspeaker programme
selector need not be at point I I. Talk-back is automatically fed to
headphones, as in rehearsal conditions. A toggle switch is available
to turn off the studio loudspeaker, if required, and when this is done
talk-back is only available on headphones.
Transmission condition is also set up and held by the signalling
red light voltage from the control room. In order to provide full
Rcuerbnathn mix-
ture switch
talk-back facilities it is usual to carry out recordings in the rehearsal
condition, using the local red light key to operate the warning
lights outside the studio and cubicle doors.
(e) Comprehensive Echo (fiverberation) Facilities
On Mark V and larger consoles, it is possible to pre-select echo on
all channels, except the Independent channel. Coarse selection
of the directlecho ratio is made by the echo mixture switch for the
channel(s) in question. Fine control of the amount of echo is
possible on the echo microphone fader. An echo cut key permits
instantaneous " make " or " break" of echo.
The chain of events may be traced by reference to Fig. 7.19 and
the following description :-
T k Echo Chain. The output of each channel fader is split into
two separate feeds by the hybrid transformer, which has two identical
and independent secondary windings. These are connected to the
direct and echo halves of the echo mixture switch (Fig. 7.20). This
is really a twochannel iider, introducing loss in the direct and echo
chains on clockwise and anti-clockwise movement respectively.
In position 5 of the switch, no attenuation appears in either chain,
and the amounts of echo and direct are equal. Switching from 5 to
I reduces echo while leaving direct at full volume.
advancing from 5 through to g leaves the echo full on and progressively attenuates the direct until at position g the direct circuit
Table 7.1
74 dB
34 dB
is broken altogether. The loss in each channel for all switch
positions is given in Table 7. 1.
The direct outputs of all switches are connected together, and taken
to the group fader. The echo outputs are taken to the echo room
loudspeaker, via echo Br and Bn amplifiers.
The echo microphone picks up this output, ptus reverberation, and
feeds it back to the group fader, via A amplher, echo microphone
fader, and the echo cut key. Howl-round of the echo chain is
prevented by the hybrid transformer.
The cross-plugging jackfield comes between the A amplifier and
the fader, as usual, so that equipment for modifying the frequency
response may be inserted. The output of the echo microphone
cannot be cross-plugged to any other channel. The independent, or
narrator's channel cannot have echo, and bypasses the group fader
into the BI amplifier.
(f) Standby Amplihs
As indicated in the simplified diagram (Fig. 7.2 I ) , standby
amplifiers are provided. There is an amplifier change-over key
for each A amplifier. Throwing one of these keys down replaces
the appropriate A amplifier by the X standby amplifier. Throwing
a key up substitutes the Y standby.
If channels are cross-plugged, the change-over key associated
with the source not the channel, must be selected, engraved studs
being provided for insertion alongside appropriate faders.
A further change-over key substitutes standby B I and B2 amplifiers for either the control (key down) or echo (key up) B amplifiers.
Since transmission talk-back (see section (d) ) uses the spare B
amplifiers, this facility is carried by faulty amplifiers if the spares
have been selected to replace faulty control or echo amplifiers.
If the cubicle loudrpeaker fails, and no headphones are available, the
key labelled " standby loudrpeaker " may be thrown, and connects the
L/S in the left-hand telephone reces direct to the HV/7.
If the moniton'ng amplijer fails, the key labelled " monitor to l i w "
may be thrown, and connects ring-main point I I direct to the
A r n w flrns
Fw. 7.21. &#&d diagram of Tjyk A Mark V
output line (Fig. 7.22). Complete failure of the monitoring amplifier
would, of course, render the programme meter inoperative.
Clean-feed Working
When a programme from a given studio includes contributions
h m remote studios or O.B. points, it is usual to feed the output of
the main studio desk to the outside sources for cueing purposes
and talk-back. In the ordinary way this cue feed will consist of the
complete programme as mixed at the main studio.
In clean-feed working this may be used by the remote studio as
a source of programme, mixing taking place at both ends. I t is
therefore essential that the remote contribution is not included,
since howl-round might result. This type of feed has also been
found useful for public address and cueing over long circuits when
the remote contributor has not wished to hear his own words on
Type A desks can have up to four outside sources, one of which is
equipped for clean-feed working. (In some exceptional studios, all
Outside sources are equipped for clean-feed working if required.)
The locking key on the left-hand panel labelled " clean feed " and
" clean-feed and TTB " gives this facility as required (Fig. 7.23).
In the centre position%f the key (normal), no clean feed is provided, communication and cueing of outside sources being effected
by means of the control line, and the remote contributor will hear
his voice coming back.
In the clean feed position of the key, used for genuine two-way
working, the output of the studio, except for the outside source
-. .-
fig. 7.23. Clean-feed working. All relays shown in "Normal" position. When clean-feed key is made relays TWA, T W B oparnta
hammis& talk-back available tu s M w and cue circuit only via 7 7 0 , STB. When clean-feed and T T B is made, TWA, T W B
dpcrata as bc$ore hansmission talk-back is auailablc to s M w and cue circuit via TTOJ STB as befms, and ako, on separate key, via CFTB,
T T 0 to clean feed
I 26
contribution, is fed back along the clean feed line to the outside source. In the rehearsal condition studio talk-back will also go
to the outside source, but in the transmission condition this is
undesirable since it would be radiated by the distant station.
In the "clean-feed with Transmission Talk-back" position of
the key the conditions are as above with the addition that it is
possible to talk to the remote contributor under transmission conditions by using the "clean-feed and talk-back" key. This hcility is
useful when the remote contributor is at a great distance away,
say in another country, when it would be undesirable for him to
hear his contribution " coming back " as in " Normal " working.
The time delay inherent in long land lines would be very codusing
for him since he would hear each syllable a fraction of time after he
had uttered it.
Points to remember in clean-feed working are:-
(I) The incoming feed is tied to a particular channel fader and must
not be cross-plugged.
(2) The incoming feed does not pass through the main hder, and
care must be taken to fade out the channel as well as the main
fader at the end of a recording or transmission. Also, since the
level of the incoming contribution cannot be boosted by means
of the main fader, it should be checked at rehearsal, and cases
of low level referred to the control room.
(3) The facility should be used only in cases where a clean feed
is required--e.g. in exchange programmes with the Continent,
the key should remain in its Normal position for
ordinary programmes involving outside sources4.S.1, O.S.2,
etc.-which can be cross-plugged in the usual way to any channel.
A number of Talks studios in London and the Regions have been
equipped with commercial desks of this type. They provide high
level mixing of four channels. The normal arrangement is shown in
Fig. 7.24-three low-level studio sources, including grams, and an
outside source channel to which one of a number of outside sources
may be switched. The faders are numbered in the diagram to
show their position on the desk, reading from left to right.
A headphone jack and " monitor " switch on the desk permits
pre-fade listening on all channels and outside source lincs-points
marked " X " in the diagram.
Fig. 7.24. Circuit of Standard Marconi Desk
TO ~ I G H T
Fig. 7-25. 7 -B Sludio [email protected] (Mark IZ)-sainpljlicd diagram
Other points of difference from the Type A Equipment may be
summarised as follows :No standby amplifiers.
Rehearsal talk-back bypasses the programme selector switch.
3. No transmission talk-back.
4. Line-up tone is usually fed through the O.S. fader (No. 2) as well
as the main fader (No. I )
5. The signalling red light from the control room holds the
equipment in transmission condition (to prevent accidental use
of talk-back) as long as it is applied, and the equipment reverts
to rehearsal condition, when the light goes out.
6. The faders cannot be removed from the panel.
When a third microphone is provided in the studio, it is wired
as one of the O.S. inputs to fader No. 2.
(Plates 7.5, 7.6 and 7.7)
A feature of the control desk is the use of standard 5& in.
long by 4) in. wide panels. It was planned, on the " building
bricks " principle, to treat each installation as an individual for the
detailed grouping of these panels. A principal feature of the equipment is its flexibility, enabling a great variety of circuit connections
to be made. For example, provision is made for clean-feed working.
Type 3 Studio Equipment-Gend
Circuit Deteils-General Purpose Desk (MarkII) (Fig. 7.25)
The Standard Type B desk will carry up to 10 channels with a
possible extension to a maximum of I 2. The facilities provided will
be taken in order.,
( a ) Source Selection (Plate 7.6)
Source to channel selection is possible, but no automatic connections exist. I t is necessary to use a double-ended cord for
every connection. Notice also that cross-plugging takes place befbre
the pre-amplifier (unlike Type A), so that exchanging channels is
an effective cure for a faulty amplifier.
Channel g is wired permanently to grams, and it is possible to
switch its output to an acoustic effects loudspeaker in the studio. A
toggle switch alongside channel 8 can be thrown to split grams
between channel 8 and 9. In this case the channel 8 cannot be used
for a microphone. The switch alongside channel g will then
connect the turntables left to it either to Direct or Acoustic. The
actual division of turntables between channels g and 8B is fixed at
the time of installation, but in a 6-turntabie studio it would usually
be 4 turntables on channel 8B and 2 on channel 9.
(b) Group Selection
There are two group faders, and all sources except the independent channel lo may be connected to either group by means of a
group switch on the channel panel. Blue and red indicator l a m p
show whether a channel has been selected to the right or left group
(c) Cue Light Selection (Plate 7.6)
Each of the green cue lights in the studio may be switched on a
cue selection panel to any or all of the cue keys which are fitted
beside every microphone fader. The operations of fading up and
cueing can thus be performed with one hand.
(d) Echo Selection
Echo is available on all channels except the Independent, by
means of the type A arrangement of individual channel selection, the
echo mixture switches being on the left of the main panel.
Standby Amplifiers
A spare amplifier, which carries transmission talk-back, may be
interchanged with either the control or echo B amplifier. Inserting
a spare A amplifier involves setting a rotary switch on the amplifier bay to the appropriate position (when the faulty and spare
amplifiers are operating in parallel), and then physically removing
the faulty amplifier.
(f) Diitortion Units
Two effects units are wired to preferred channels, one of which
will always be grams, so that bass cut and top cut may be inserted
by switches on the panel.
(g) Public Address Selection
Where public address loudspeakers are required, a special panel
of switches selects channels I to g to be fed to the P.A. amplifier.
(h) Pre-Jadc Facilities
Outside sources are provided and pre-fade listening to these
sources is selected by push-buttons. The panel which carries the
pre-fade buttons also gives flexible cue speak Lcilities. The control
line normally carries cue programme and talk-back to the outside
source. This is replaced by the control tekphonc when the speak
is depressed.
(i) Push-button Selection
Buttons are provided, as with Type A, for rehearsal, line-up,
and transmission. No locking into transmission condition by the
red light voltage occurs.
(j ) Clean-feed Facififiw
The incoming contribution is fed to the blue group and all studio
sources must be fed to the red group. The blue group control then
acts as the fader for the source to which clean feed is sent. Clean
feed, and clean feed with TTB, are provided as with Type A.
Tplhs d special Desks
(a) The Talks and Dis-om
version of the Type B desk (Mark I)
is accommodated on a panel of the same dimensions as the centre
panel of the General Purpose desk. The bottom row of five channels
are, from left to right, microphone I , microphone 2, incoming
contribution, tape and gramophone, The incoming contribution
channel is for use in clean-feed working only. The middle row of
panels carries the main control, and two outside source faders,
extending on certain installations to a maximum of four.
In general, the facilities are the same as with the General Purpose
desk, except that no echo, group, or independent channels are
(b) The S ' a l versions of the desk (Mark 111) have been designed
to suit individual large studios. A typical arrangement which is
regarded as a working maximum consists of I I main channels on
the centre panel, together with the main control, group, independent and echo faders, plus a further pair of independent channels,
and 16sub-channels mounted as two +channel mixers on either side
of the centre panel. One or,more of these mixers may be plugged
to any of the I I main channels, or 3 Independent channels, for
grouping purposes. The maximum number of sources available
simultaneously is 28. Amongst other facilities found in the Mark
I11 equipment are the following:I. Switchable cue lights up to a maximum of I I
2. An alternative echo source, when available, necessitating altering the Independent channel on the centre panel to " Echo 2 ",
and provision of a switch alongside each Echo Mixture Switch to
select the two echo rooms or machines
3. Cubicle L/S may be switched to monitor the public addresse.g. during the sending of " line-up " tone when an audience
warm up " is in progress
4. Public address switches for all eleven main channels, and I
Independent channel-having an up position giving 10dB boost.
I t is also possible to feed echo I to P.A.
5. Distort units pluggable to any channel
6. Split gram outputs pluggable to any channel
Detailed descriptions of particular installations are outside the
scope of this handbook, but the B.I mixer suite in Broadcasting
House incorporates a number of interesting features which may
indicate future trends in design, and therefore a short description
may be usell.
As the photograph, Plate 7.8 shows, the mixer desk is based on the
Type B studio equipment, with special arrangements to suit the
primary function for which the suite was designed-namely, the
mixing and direction of large-scale broadcasts involving a number
of outside sources or studios.
The principal mixing controls are situated on the centre panel of
the mixer desk, with the main control hder on the left-hand panel.
A programme meter is provided on each of these panels, in case a
second studio manager becomes necessary to control the overall
volume. There are two goup faders a t the top of the centre panel.
Each controls the row of three channel faders immediatelv below it.
coloured red and blue respectively. Each of these chainel faders
may be used to control any one of four different sources by pushbutton selection. All sour&s are plugged to the channel inbits on
the control position in the engineer's cubicle, in accordance with
the layout requested by the studio manager, and selection of the
required source on each four-source fader is achieved by pressing
the appropriate push-button on the left-hand side of the fader. When
connection is made of the source to the input of the fader, the
corresponding lamp above the fader will light.
This selection of a source is possible only when the channel is
fsded out. Thus, if source L.G.IA is k l c d up (the first source on
the first fader on the left group), the button ass0ciated with L.G.
B, C, or D may be depressed, but the connection to the input of the
fader will not change over until the fader is hded out. This system
of pre-selection is designed to prevent accidental change-overs, and
to facilitate the change-over when consecutive programme items come
up on the same fader. When the running order is known in advance,
of course, sources which follow consecutively should be plugged to
different faders, and when the total number of sources is 10 or less,
each should be allotted a separate fader.
Four independent ciumrul faders, designated W, X, Y, and 2,
occupy the bottom row on the centre panel. The desk as a whole
can therefore be used with up to 28 sources, although only 10
channel faders are provided.
BB Mix- Local Output
The B.I mixer room includes an announcer's desk of the type
employed in B.H. Continuity studios. The output of this desk can
be plugged as a source to the mixer desk, and includes turntables
for 78 r.p.m. and microgroove discs, and a microphone suspended
over the announcer's position. This is intended to link composite
programmes through the mixer, or can be used as a standby continuity suite. The outputs of the three 78 r.p.m. desks and the
L.P. desk also appear as part of the mixer output.
An auxiliary microphone socket and fader on the right-hand
side of the mixer control desk are intended for use with a lip microphone. I t is important not to fade up the announcer's desk
microphone simultaneously with the lip microphone, since the bass
filter circuit of the lip microphone will deteriorate the quality from
the other. When either microphone is faded up, the mixer room
loudspeaker is automatically cut or dimmed in the ordinary way.
Local Tape Facilities
A special jackfield is provided on the left-hand wall, whose 28
jacks, suitably coloured and labelled, are connected in parallel with
the 28 channels on the desk. By connecting the appropriate jack
to a recording machine, it is thus possible to continue radiating from
any channel, while recording from another for future reproduction.
The output of one or more tape machines set up in the mixer may
be plugged as a source to any channel of the desk. The correct
procedure for this is to plug the output of the tape machine to the
rekro. jack, where it can be picked up on the Control Position, and
P a 7.1a .
Peak prgrammc meter
Typc A control cabinet
Plate 7.3.
Type .4 comole
Platt 7.4. Studio control console .\larconi Type BD.5q.j
Plate 7.5. General purpose desk- Type B equipment
Plate 7.6.
Type B equipment-pre-selection cabinet, showing cue-light switches on the left,
and source/channeljacks on the right
Plate 7 . 7 T y p e B equipment-arnpl$er rack. showing two groups of
10 arnpliJicrs. ~tandbe arnp11Jier and changeow swikh, and maim unit
Plate 7.8. Control desk in Mixer Suite B . I ., Broadcasting House
Plate 7.9. Special desk for BBC studio in Paris
Plate 7.10. Special desk for Coven&Garden Opera House
plugged to any channel as required. Note also that material on
any control room circuit not booked as a source to the mixer may
be recorded in the mixer room by requesting that it be fed to the
'' miscellaneous record "jack.
Outside Sources
On the right-hand panel are situated 28 individual talk-backlprefade keys, which provide comprehensive facilities for passing directions to individual contributors. The keys are arranged in groups,
and coloured to correspond with the arrangement of channels on
the centre panel.
In the normal position of each key, cue programme is fed to the
source, there being a cue programme selector which, on point 11,
feeds the mixer desk output, and a change-over key corresponding
to the " white " key alongside the mixer programme selector. Note
that in the case of the independent channels W, X, Y, and Z, this
cue programme will normally take the form of a cleanfeed on point I I.
In the up position of each key (locking) pre-fade listening on the
associated source is provided on the desk headphones-and also on
the mixer room. loudspeaker during rehearsal or recording.
In the down position of each key (non-locking), the cue feed to the
particular source is replaced by talk-back, and the desk headphones
are switched to pre-fade listening. During rehearsal, the mixer
room louds~eakeris also switched to re-fade in this condition.
but at reduced volume to avoid howl-round. During transsmision,
pre-fade may be switched to the L/S by the transmission Prefade Listen switch. A s~ecialwhite kev on the left-hand an el
enables the studio manager to switch his headphones to programme,
even when the producer may be pre-fading on the loudspeaker.
Onlv one of the individual talk-backhe-fade kevs should be used
at a time. Two " master " talk-back k'eys on the Lentre panel give
simultaneous talk-back to all sources on the two group faders, but
the only means of speaking down the clean feeds to sources W,
X, Y, and Z is via the individual keys. In a case where only clean
feed is required without inadvertent use of talk-back-the genuine
" multi-way " programme-this must be booked specially and
Control Room will render clean-feed talk-back inoperative on
To avoid the automatic dimming of the mixer room loudspeaker,
which accompanies use of the ordinary talk-back microphone, an
auxiliary socket is fitted on the skirting-board for connection of a
lip microphone. When this connection is made, the lip microphone
replaces the ordinary talk-back microphone, and the loudspeaker is
not dimmed when talk-back is in use.
7.6.5. Green Cue Lights
There is a green key to the right of each of the Independent
channel faders. Operation of these keys energises the " Master
Cue " relays of any studio in Broadcasting House which is plugged
as a source on the corresponding channel. In addition, a green key
alongside each group fader will operate the master cue circuit of
any Broadcasting House studio on that p u p whose channel is
faded up.
7.6.6. Stcldio Red Lights
When the red light is applied to the mixer for transmission or
recordings, the red light is also operated for any studio in Broadcasting House, Maida Vale, or I & I A Portland Place, which is
connected to a channel fader. In the case of four-source faders, the
studio red light is applied when the corresponding indicator lamp
is lit. This means that deliberate selection of the source will
apply the remote red light when required-ither
for cueing a studio
or for starting a reproduction from any tape reproducing channel
which has been selected as a source on the mixer control desk.
7.6.7. Clean-feed Facilities
As mentioned above, the normal feed of cue programme to
sources W, X, Y, and Z takes the form of a clean feed. Thus, with
the special cue programme selector switch at point 1 1 (or the
associated change-over key hum), each outside source receives a
clean feed of any of the other 27 sources which are faded up-on
the control line for O.B.s, etc., and on line 10for local studios.
7.6.8. Commdcations with Enginccis Cubicle
A. talk-back unit providing two-way communication with the
engineer's cubicle is situated on the right-hand panel of the desk.
In addition to the special versions of standard studio desks
described above, it is necessary, from time to time, to produce
specially designed desks to fulfil a particular function. Two
examples will be described.
BBC Studio in Parh
Plate 7.9 shows the desk designed for the BBC studio in Paris.
Since most of the work of this studio is transmitted over land line to
London, special talk-back and cueing arrangements are included.
The two tape machines inset into the left-hand " wing " of the desk
can be remotely controlled from the desk itself. Replay, record,
and fast spool facilities are provided in this way.
The desk provides means of controlling a second, smaller studio
in addition to the main one when, for instance, a narrator is necessary for a feature programme.
BBC Equipment at C-t
Garden Opera House (Plate 7. to)
The listening room a t Covent Garden is high up at the back of the
gallery, and the acute viewing angle to the stage made a " flat "
design essential. Quadrant faders were used, and various facilities
were provided. In addition to the main slung orchestral microphone and two stage microphones mounted in the " float " there
is provision for orchestral microphones in the pit, and microphones
for " off stage " effects. A channel for artificial reverberation is
provided. Two independent channels for narration and announcements are wired to extensions in one of the theatre boxes. Talkback and cueing to the announcers and stage are available.
A somewhat unusual facility is provided in that it is possible
to obtain hvo feeds to line, one the normal domestic feed including
local announcements, the other being music only, for transmission
to, say, a foreign radio organisation who can then add their own
announcements. This latter is also termed a " clean feed ", a
slightly different use of the term to that mentioned earlier.
For all normal applications, studio and transmission equipment
is designed so that it has a flat frequency response within the audible
spectrum. There are occasions, however, when it may be necessary
to introduce variations in this response, either for deliberate special
effects, or to correct acoustical or recording quality. Some
examples of this will now be given.
MicTophone Comcti011 Units
These are primarily designed to offset the increase in low frequency response that takes place on close speech-particularly
1 3 ~
with ribbon microphones. (See Chapter 5.) The circuit and
frequency characteristics of a typical M.C.U. are shown in Fig. 4 . 1 4 .
A close technique, and therefore a Bass Correction Unit will be
necessary in the following circumstances :(a) In small rectangular studios boomy quality is often caused by
dimensional resonances, despite extensive acoustic treatment.
Introducing some degree of bass correction permits a closer
technique, and at the same time reduces the studio colorationwhich is often at low frequencies.
(b) In large orchestral studios, a normal working distance for
announcements may give unduly reverberant speech. Occasionally, therefore, the announcements may be made fiom less
than 2 ft away, and bass correction used. It is desirable, at the
same time, to ensure that the announcer sounds " in the same
place " as the orchestra, etc.4.e. in the same " acoustic ".
(c) For dance band vocalist., a very close technique has become
the rule, and a number of portable bass correction units have
been made available for London studios. They are labelled:" Bass Cut for Ribbon Microphone only at less than two feet
Portable Effects Unit PEU/x
This unit provides five amounts of Top Cut and four of Bass Cut,
selected by separate switches. As might be expected, the overall
Fig. 7.26. [email protected] curves of PEUIr
level depends to some extent on the degree of cut, and a careful note
of the control settings should be made at rehearsal.
The PEU/I is recommended when fixed amounts of distortion
are required for special effects. I t is much to be preferred to
fastening a baffle on to the microphone. The latter procedure
may succeed for one voice in a fixed position, but gives uneven and
unpredictable results in more complicated layouts, or where movements take place. It can never be used for serious music.
An idea of the distortion to expect from the PEU/I may be
obtained from the response curves shown in Fig. 7.26.
Variable Correction Unit VCU/xA
I t is occasionally desired to transmit programmes which have
been recorded with a different characteristic from that used by
the BBC. Accordingly, the VCU/IA is fitted in certain channels,
and may be adjusted aurally for best transmitted quality. For example, a Top Cut at 6 kc/s might be considered desirable on an
overseas despatch, or on a Home programme which contained a
number of very old disc inserts.
There are four correction circuits which may be used independently or in any combination:-
Bass Increase-decreaseto vary the general shape of the
2. Top Increase-decrease
response curves
3. Top Cut-to suppress surface noise etc. (4,5,6, 7, or 8 kc/s)
4. Bass Dip-to suppress mains hum (50, 60, 100, or 120 c/s)
Occasional use of VCU/IA in studios has been made to eliminate
variations in the recording characteristics of some commercial
micrograove recordings.
Gxn Filter-Fitted om Turntable Desks TD/7
The BBC uses a different frequency characteristic for 78 r.p.m.
discs than do commercial companies. For this reason, and so as to
take full advantage of the relative freedom from surface noise which
is possible in direct recordings, turntable desks include a two-position
filter switch. This allows both types of disc to be reproduced, the
overall levels being matched up (f4 dB) at the same time. The
effect on a direct recording of the wrong key setting is loss of top;
and for a gramophone record produces excessive surface noise.
The materials used in recent BBC pressings can result in a
freedom from surface noise which is comparable with that of a direct
recording. New pressing--especially pressings of music--should
therefore be played in the " Direct " position unless surface noise
is particularly noticeable.
1 3 ~
Qpad" Quality CoPtrol Unit
This is a commercially manufactured domestic pre-amplifier/
tone control unit, which has been modified for use in BBC
studios. As modified, it has zero dB gain, and 6oo ohm input and
output impedances. The volume control is rendered inoperative.
The first BBC studio application of this instrument was in
gramophone studios, where it was arranged so as to be available on
Fig. 7.17. Quod qualie control d .
Charactnirtics of ( a ) bass and treble
controls, ( 6 ) flh
TD/7 outputs only. Its purpose was then to apply filtering to 78
r.p.m. records to remove surface noise, excessive high frequency
emphasis, and distortion. A key was provided to cut the apparatus
in and out of circuit. With the advent of the R P ~ I Ithese
have now been connected in such a way that they can be used on all
types of gram desk, and so obviously become available for use on
fine-groove discs.
I t must be stressed that the use of the Quad should be confined
to discs where there are faulty conditions to correct. Any attempt
by a broadcasting organisation to alter the balance or general sound
of a commercial disc is liable to incur the displeasure of the manufacturers !
More recently, these units have been used experimentally in
microphone circuits, mainly in light music and light entertainment
programmes, both in order to achieve better separation between
sections in a multi-microphone balance, and for special musical
The facilities provided are as follows. Bass lift and cue at a rate
of 6 dB/octave, with frequency of turnover variable, such that at
maximum boost, the turnover is approximately I kc/s, with a
maximum variation of f 10 dB at IOO c/s. Treble lift and cut,
turning over at I kc/s, with a maximum of f 15 dB at ~o,ooocis.
4 on these two
I t must be noted that the calibration of - 4, o,
controls is purely arbitrary, and not a dB scale (Fig. 7.27 (a).
A low pass filter is also available, having a choice of three turnover frequencies, 5, 7, and 10 kc/s, the slope of the roll-off being
continuously variable between o and 50 dB/octave. In this case
the control is actually calibrated in dB/octave (Fig. 7.27 (b).
The push-buttons on the unit are normally inoperative for
BBC use, but an experimental version has been produced where
these are used to switch into circuit either of two presence filters,
giving 2,4, or 6 dB rise at 2-8 and 6 kc/s.
A temporary rig capable of easy dismantling, yet providing
elaborate facilities when required, is the general description of a
set of O.B. equipment. Only the basic layout for the OBA/8
and the OBA/g equipments will be described. The more complicated programmes call for an extension of these layouts in an
almost,inhite variety of combinations.
8.1.I. OBA/8 Equipment
This equipment consists of the mixer Mx/18, the amplifier
OBA/8, and its associated power supply, together with a monitoring
loudspeaker and amplifier (Plate 8. I and Fig. 8. I).
(a) Mixer MXIx8 (Fig.
Each channel of this mixer is a balanced potentiometer introducing attenuation in 2 dB steps, with 3 dB increments on the last
four studs. A study of the diagram shows that each channel
" sees " the output with the other three channels in parallel. When
the other three channels are faded out, Microphone I , for example,
is correctly matched into the amplifier (300 ohms). But fading up a
second channel (with microphone or grams attached) upsets the
matching, and a drop in level of as much as 3.5 dB occurs-an
amount which may call for compensation on the main control
Fading up further channels results in more loss, in smaller increments. Three extra channels, for example, reduce the level by
about 8 dB.
In the diagram a high resistance or " static leak " is connected
from the end of the resistor to the of stud in order to preserve
electrical continuity and prevent clicks due to building up of
Plate 8.I .
Typical arrangement of O B A / 8 equipment
Plate 8.2. OB.419 trolity
Platc 8 . 3 .\lobile Control Room
Plate 8.4. .%!/-operated O.B. eqripment
charges on the contacts. Only one of these is shown for simplicity,
though more may be used.
(b) Outside Broa&mt AmpliJier OBA/8
This is a two-stage amplifier, incorporating a gain control and the
necessary valve circuits to feed a peak programme meter. The input
impedance of the amplifier is 300 ohms, and the maximum gain
is 91 dB (control a t 35). Fading-down introduces attentuation at
Fig. 8.1. B k k
diagram of
O.B. rig and linc co~dcsions
2 dB per stop from 35 to I I , and 3 dB from I I to I, with fade-out
at zero. At an average setting of 25, the gain is about 70 dB, thus
raising the output of microphones or grams to about zero level.
An output attentwtor introduces o, 4, or 8 dB attenuation, giving
output levels (P.P.M. reading 4) of 4,0, or - 4 dB respectively.
The output circuit is loaded in the receiving control room in such
a way (239 ohms) as to make this sending level correspond to zero
voltage level. (For definition see Appendix.)
(c) Calibration of Programme Meter
Regular calibration is necessary and in order to do this the
following steps should be taken :(I)
Allow 10 minutes for amplifier to warm up, after switching on.
Fade out main control.
Unplug loudspeaker if LSM jack is used.
Adjust zero by means of the control. marked " Adj. zero
(5) Throw the calibration switch to CAL.
(6) Adjust meter to read 4 by means of the control marked " Adj.
(7) Return switch to Normal and re-plug the loudspeaker.
For O.B.s, a minimum of two amplifiers and mains units are
installed as a precaution against breakdown, and two sets of
batteries are also connected in case of failure of the mains supply.
Provision is made for rapid interchange of the programme and
control lines to the studio centre.
The block diagram shows the mixer M X / I ~connected to the
amplifiers in parallel. The input switch of the spare amplifier is
at of. The output jacks of the amplifiers are also paralleled, using
double ended cords. The normal condition is main amplifier
output to Line I, telephone to Line 2, this being arranged by
suitable positioning of the output keys.
(d) Amgljer
In the event of failure of the main amplifier, or its mains unit,
the input switch on the spare is thrown to on (the gain control
being at a suitable setting) and the other switched oJ.
It is only
necessary then to transfer the loudspeaker cord to the LSM jack
on the spare amplifier.
During this change-over, the amplifiers are operating in parallel
for a second or so. I t is therefore usual to test during the installation that the feeds from the mixer are " in phase ", otherwise
cancellation would take place in the output to line.
(e) Line Change-over
If the programme line-line
I--develops a fault, the control
room will telephone on Line 2 requesting a line change-over. After
a pre-arranged interval-say, 15 seconds-the Line 2 change-over
key should be thrown from TELE to AMP, and after a pause,
Line I key may be moved from AMP to TELE. Exactly on the
15-second cue, the control room will have changed over the termination of Lines I and 2 by re-plugging, and will then telephone to
confirm satisfactory results.
OBA/g Equipment (Plate 8.2)
A considerable amount of auxiliary equipment is necessary with
the OBA/8 rig, and the OBA/g was developed in 1950 to simplify
transportation, ctc.
The equipment fits on to a trolley, which can be carried in a
large car. The illustration shows the front of the trolley, with
mixer MX/2g, spare and main OBA/g, distribution unit, LSM/g
and supply unit. On the back of the trolley are three drums of
microphone cable which can be run out without dismantling.
The mixer unit MX/29 provides for four low level input channels
of 300 ohms or 30 ohms impedance. Attenuators are also included
in case one high level source is to be mixed.
The amplifier OBAlg provides similar facilities to the OBA/8,
except for output line switching.
The dktribution unit DU/I provides flexible connection of programme and telephone outputs to the studio centre, as well as to
on-site commentators, etc.
The LSMIg unit feeds the monitoring loudspeakers, and also
supplies trapvalve feeds to public address systeins, and other
destinations as required.
The *ply unit SUP16 contains both batteries and a mains unit.
The normal transmission arrangement is main amplifier mains
operated, and spare amplifier connected to the batteries.
Terminadon of Lines at 0.B. Point
At permanent O.B. points, the microphone extension terminals
and the P.O. Exchange lines should all be clearly numbered and
labelled. The numbering of the P.O. lines should correspond with
LlNE 1
Fig. 8.2. Looping of finLs
the local end numbering in the BBC control room through
which the O.B. is routed.
At ImrpMmp O.B. points, the P.O. line terminations should be
clearly labelled as soon as they can be checked over with the
control room concerned.
When an O.B. point is left unattended, it is standard practice to
leave the lines looped together in a certain way, to enable conclusive
t a t s to be camed out &om the BBC control room or P.O.
exchange. The method of looping is A-A and BB, as shown in
Fig. 8.2.
When an odd number of circuits exists-say, one music and
control plus a cue or spare music l i n e t h e odd line is looped -4-B.
When an O.B. point has its exchange lines connected through to a
BBC control room, the lines must be left looped in accordance
with the above if the engineers are absent from the O.B. point for
periods exceeding half-an-hour.
Tests on Lines from an 05. Poht
I t is the duty of the senior O.B. engineer to ensure as soon as
possible after arrival a t an O.B. point that tests are carried out on
the exchange lines.
The lines may be connected direct to the local BBC control
room (see Fig. 8.3 (a) ), or to the P.O. exchange for extension to
trunks or other circuits on to BBC networks (see Fig. 8.3 (b) ).
The BBC control room in (a) will probably have connected a
telephone ringing indicator to one of the circuits in anticipation of
Fig. 8.3. Line co~ccliau
the O.B., and communication by field telephone should be possible
from the outset. If connection cannot be established, an outride
telephone call should be made without delay to establish the cause
of the fault.
In (b) an ordinary telephone call to the local P.O. exchange
should be made to arrange the tests.
The tests consist of speaking on each circuit in turn, and cooperation in D.C. tests as required. In the D.C. tests, a s h t
circuit is applied to each line in turn, to allow the remote engineer to
measure the resistance of the line. Then the ends are left open
while insulation between the two " legs " and earth are tested.
A jwogrmme test should also be originated from all O.B. points,
to enable the control room to judge the incoming quality. Part of
the rehearsal may be used, or if this is not possible, speech on one
of the microphones installed for the O.B., or special one installed
for the purpose near the apparatus. This programme test is particularly important where the control circuit is unsuitable for music, as
it will disclose cases where the music and control circuits have been
inadvertently crossed.
It will be appreciated that during the final period before the
O.B. starts, the fullest co-operation must be extended by both ends.
This will consist in general of answering the telephone promptly,
or warning the other end if for any reason this will not be possible
during any particular period.
This is a small, self-contained equipment for use by commentators on occasions where there is no need for a full scale outside
broadcast rig, and where no engineer is necessary (Plate 8.4).
The equipment consists of a suitcase-mounted amplifier with
line switching equipment to enable the necessary pre-transmission
tests to be made. Talk-back and cueing.from the receiving control
room or studio are possible on headphones, and a miniature receiver
is provided for transmission cue.
The amplifier is battery operated, transistorised, and the microphone used is the Acos Mic.39.
In studios where an audience is admitted, and more particularly in studios where Light Entertainment programmes are
taking place, it is necessary, in order that the audience may hear
the broadcast more satisfactorily, that a public address system
should be installed. P.A. equipment is also used under Outside
Broadcast conditions, when if an existing installation is present this
must be linked to the broadcasting equipment, and if not a complete
system must be specially rigged for the occasion. The use of public
address introduces a number of problems and in this chapter some
of the difficulties will be described.
In any P.A. system the basic requirements are obvioudy a microphone, a power amplifier and one or more loudspeaken. The
microphone may be a separate instrument or, and this is more likely,
be one or more of the microphones used for broadcasting purposes,
the output of these microphones being split so that they can feed
both the broadcast and public address systems.
The effectiveness of a P.A. installation may be judged under the
following headings :(a) Loudness
(b) Quality
(c) Instability
(d) Sense of direction
Correct loudness at all points in the auditorium is not easy to
achieve. If there is sufficient volume of sound to satisfy the back
rows of seats, there is a danger of excessive loudnw for people near
the loudspeakers. The cure for this is to mount the loudspeakers
high, so that the nearby audience is on the fringe of the loudspeaker's polar diagram. Since these people are near the stage,
direct sound will ensure that they hear well. -The back rows of
seats receive a useful reinforcement of sound by reflection from the
rear wall (see arrow in Fig. g . ~ ) ,but angling of the loudspeakers
Fig. g.r.
Basic Public Address system
must be such as to prevent these reflections reaching seats further
than 20 ft from the rear wall, or else echoes will be introduced.
The quality of the reproduction is at best a compromise, bearing
in mind the varying directivity of loudspeakers, and the overlapping
areas covered by the direct sound and the loudspeakers. For best
intelligibility there should be some attenuation of the bass frequencies, and this can be done in the P.A. chain, either by using
loudspeakers with inefficient bass response or by means of deliberate
I t follows that deterioration of broadcast quality will result if
much of the P.A. radiation is allowed to spill on to the microphones-either stage microphones, or those used to pick up audience
reaction-and the siting of microphones and loudspeakers must be
regarded as an inter-dependent operation.
Correct phasing of the loudspeakers is important. To check this,
it is usual to match up the volume from each loudspeaker individually, listening at some point equidistant from them, and then to
switch in both together and arrange the terminal connections for
1 4 ~
an improvement in volume. This test is better carried out on low
frequency tone, or even mains hum.
Instability may occur when the microphone feeding PA. is itself
within range of the loudspeakers. There is, in such circumstances,
a danger of the sounds being repeatedly amplified-until a howlround takes place--causing loud oscillations if it is permitted to
buiId up. Incipient cases may not bring about continuous howlround, but the resulting distortion is enough to mar programme.
The solution again lies in correct siting of the microphones and
loudspeakers, and correct setting of the P.A. gain control.
The relative phase of the microphone to the loudspeakers will
often affect the onset of howl-round. This may be checked by
fading-up to the point of instability, and reversing the microphone
Fig. 9.2. Siting of Mspakrrs lo ensure good "sense
of direction''
terminals--or the microphone itself in the case of ribbons-to
discover the safer condition. When roving microphones are used,
the problem is made more difficult.
Other things being equal, our sense of direction, when listening to
more than one source of a given programme, works in such a way
that we associate direction with the source which reuches ur fist.
Within certain limits, this remains true even if the later sounds are
stronger than the first. In P.A. work it is desirable, therefore, to
arrange that the path from loudspeakers to the audience shall always
exceed the " direct" path. In Fig. 9.2, for example, two loudspeakers, P and Q, are directed at opposite diagonal corners. PB
exceeds AB, and provided loudspeaker Q is angled sufficiently to
Fig. 9.3. Typicalpolar diagram of line source at high freQuntcies
miss the listener at B, the sounds will appear to come from the
actor at A. The worst conditions exist for people near the two
loudspeakers, but some improvement is often possible if the loudspeakers are placed high (as recommended in Section 9. I. I. above),
so that the sound heard is almost entirely " live
While stressing that the reinforcement path should not exceed
the direct path, it is necessary to say that they should not differ by
more than about 20 ft, since beyond this distance a discrete echo
becomes apparent.
The loudspeakers commonly used for P.A. can be divided into
three groups, arranged in increasing order of directivity, as follows:
(a) ordinary cone loudspeakers
(bj horn loudspeakers
(c) line source loudspeakers
-4 single horn loudrpeakcl is more effective than a cone loudspeaker,
since its narrower polar diagram permits more accurate direction
of the sounds at the audience, with consequently less spill on to
microphones. Furthermore, the falling off of intensity along the
sound path is less steep, giving more even loud.ness to near and
distant listeners.
The line source, or '' column ",loudspeaker carries the " focusing "
process a stage further. I t comprises an array of loudspeakers
mounted in line in a unit 6 ft or more in length. The radiation
of sounds-particularly at high and middle frequencies-& directed
along the plane at right angles to the column, in a fan-like manner
(Fig. 9.3). If the unit is set up vertically, as is most usual, unwanted
and wasteful radiation of sounds upwards, or downwards on to the
heads of nearby listeners, is reduced-with a desirable saving in the
electrical power necessary to cover a given auditorium. It is
important to note, however, that no particular directional effect is
achieved in the horizontal plane in this case, and if this is required,
the coIumn shouId be tilted, or even set up horizontally.
A recent development in public address equipment has been the
result of some research in the United States, and this equipment has
now been used with considerable success in some BBC studios.
In this equipment an electronic device is inserted between the
microphone and the loudspeaker equipment, and this device shifts
all frequencies passing through the system by a small amount,
usually of the order of 5 cis. The object of this is that at any
given instant the direct sound arriving at the P.A. microphones
will be at a different frequency (by 5 CIS),from the sound arriving
at the P.A. loudspeakers, so there is less likelihood that the system
will howl. In practice a howl point is eventually reached, but an
im~rovementin P.A. loudness of u~wardsof 6 dB is normally
obiained. In conditions of high ambient noise this improvemen;
may not be apparently as great as this, since the background noise is
also amplified by the P.A. system and so the ratio of programme to
background noise may not be significantly improved. This, of
course, happens with ordinary P.A. systems but the added gain
with frequency shift may make the effect more noticeable.
If public address is to be achieved in very large halls, or in the
open air, members of the audience may be at a great distance from
the original sound source, and it will be necessary to place a series of
loudspeakers at various points between the back row of the audience
and the sound source. If these loudspeakers are connected normally to a P.A. system, the time taken for the sound to reach the
audience from the original sound source will be much longer than
that from the nearest loudspeaker, and so the sound will appear to
come not from the original source, but from the loudspeaker itself.
By a system of progressive delay from the original source to the most
distant loudspeaker this effect can be minimised in order to preserve
the condition that the original sound source must be heard first, if
only rather weakly.
It may be required, under some circumstances, to reproduce
stereophonic recordings introduced by speech fiom a stage, to a
large audience. This is not normally successful if only one pair
of loudspeakers is used. The situation can be improved, and a good
stereophonic effect achieved over the greater part of the audience
area, if a number of pairs of loudspeakers, arranged down the
Fig. 9.4. Stercqbhonic P.A. instcrllation
sides of the hall are used. These loudspeakers must all be in phase
with one another and each pair must be correctly balanced in order
that a monophonic signal, fed equally into the two halves of the
system, shall produce a centre image. Once each pair has been
balanced in this way it is necessary to adjust the relative volume
levels of each pair of loudspeakers with respect to the preceding and
following pairs, such that the audience at any given point in the
sitting area will only hear the pair of loudspeakers which are nearest
to them (Fig. 9.4).
.houncements can be fed into this system in two ways. The
simplest form, if only one person is speaking, is to use a monophonic microphone and to connect its output equally to the two
on PA
Fig. 9.5. P.A. in OBAI8 rig
Fig. $6. P.A. in Typc A 6puipmmt
stereophonic channels. Alternatively, if two or more persons are
speaking it may be possible, under some acoustic conditions, to
achieve a fair stereophonic effect by using a stereo P.A. microphone.
The success of the reproduction of stereophonic material in this
way cannot be predicted, but is more likely if the general acoustic
is dead. However, moderate success has been achieved in a
number of halls having widely different acoustic conditions.
The use of hybrid transformers to derive a P.A. feed from each
programme channel is also described in Chapter 7.
In OBA/8 installations, there is no special provision for P.A., and
separate hybrid transformers are simply wired at low level at the
input to the mixer, or sometimes between the mixer and the
OBA/8. Fig. 9.5 shows an alternative arrangement using a 70 dB
attenuator, and no hybrids.
With the OBAlg, the same methods are used, with the addition
that, in cases when the complete programme is to be fed to P.A..,
the trapvalve amplifier supplies a feed at about zero level for the
With Type A Equi)mmt, P.A. circuitry is included only in
" audience " studios, and takes the form shown in Fig. 9.6.
The switches SW are supplied in a few studios such as the Concert
Hall, and enable the studio manager to select sources to P.A. at will.
In other studios, selection of channels to P.A. is made on the
engineering control position, and often hybrid transformers are used
in place of attenuators. In all Type A installations the P.A. level is
dependent on the setting of both the channel fader and the volume
control on the P.A. amplifier. Another point to notice is that fading
out the group or main faders will not normally remove the feed to
P.A. if the chnnel fader is open. In certain studios, however, complete fade out of group faders cuts the feed of that group to P.A.
With Type B Equipment, switches are fitted on the control desk for
all channels which can be fed to P.A. The up position of these
switches gives a 10 dB boost to the P.A. level from each channel.
On Mark I11 desks there is an auxiliary fader giving overriding
control of the volume of P.A.
It should be noted that on all the above circuits the various
inputs to Public Address are derived from channels carrying contributions to the actual programme itself. However, this paragraph
would be incomplete without mention of the simplest arrangement of
all-namely, the use of a self-contained P.A. circuit. In this, a
separate P.A. microphone is set up alongside the broadcast microphone (or sometimes fastened to the same stand) and allows P.A.
volume to remain quite independent of mixing levels on the broadcast programme.
A number of commercial loudspeaker amplifier combinations
have been used by the BBC. At present the amplifiers most
commonly employed for P.A. work are manufactured by Pamphonic Reproducers Limited. They have a maximum power
output of 10,25, or 50 watts, and will operate either from a balanced
input at microphone level, or balanced or unbalanced inputs at
zero level. They have output connections suitable for driving line
source units or low impedance loudspeakers direct.
The loudspeakers most commonly employed are Pamphonic
6-ft or 8-ft line source units, using the principle outlined above.
To give equivalent directivity at high and low frequencies, these
units contain two lines of loudspeakers-bass and treble--with the
necessary cross-over circuits. In studio installations it is usual to
set up two columns to cover the floor area, and a further two for
the gallery as necessary.
Other loudspeakers used in Outside Broadcasts, are I 2-, 10- and
8-in. cone units mounted in box cabinets, the latter being strung out
in relatively large numbers to give low volume coverage ofaudiences.
recording in its present form is a development from a system
which grew up in Germany during the last war, but the principle
is a relatively old one, indeed it was known at the time the first
disc recordings were made at the turn of the century. In the
early days, the recording medium was a thin steel tape and recording machines employing this were in use by the BBC until shortly
after the war (Plate 10.1). This system suffered from a relatively
poor frequency response, at least by modern standards, and a
fairly high level of background noise. Other difficulties included
the necessity for a welding technique to repair a broken tape or
to make joints during editing.
The development of the " Magnetophon " tape recorder in
Germany during the war, using a plastic tape coated or impregnated with magnetic material, made high quality recording
possible, and the discovery that a high frequency bias applied
during the recording process would keep the distortion to a very
low figure paved the way to subsequent developments.
The tape recording and reproducing machine consists of a tape
transport SySm which will move the tape at a constant linear speed
across three tape heads. The tint of these is the erase head, the
function of which is to ensure that the tape is wmpletely demagnetised. The second head is the recording head, which is fed with the
programme currents which are to be recorded. The third head
is the reprodtrcing Ircad, used to monitor the tape whilst recording and
to reproduce the tape when required (Fig. 10.I).
The recording head is fed from a record amplifim which contains
part of the compensation or cgualucrtidn necessary to produce an
overall " flat " frequency characteristic. The reproducing head
1 5 ~
feeds a reproducing ~ " p i t f and
k this contains the rest of the necessary
equalisation. The erase head is fed a t a high alternating frequency,
usually between 60 kc/s and roo kc/s, from the bias oscillator, and
this oscillator, as its name implies, also feeds the high frequency
bias to the record head (Fig. I 0.2).
In the domestic tape recorder, one amplifier commonly taka on
the function of either record or reproduce, the necessary changeover of equalisers and connection being made by the record/play
Fig. r0.r. Diagram of the rhrcc kaa3 in a tap reawdcr
switch. Under these conditions a single record/play head is used,
thus saving considerably on the cost of the equipment. Of course,
under these conditions no monitoring from the tape during recording
is possible and the recording cannot be heard until after it has
ended, the tape rewound, and the equipment switched to play.
Whilst this is perfectly satisfactory for many domestic applications,
it is not practicable for a broadcasting system since it is imperative
Fig. ro.2. General s c h i c of &@ recorder
,Cfarro~li-Stillesteel taFe recorder
E..\.i.Z. recorder TRlgo showing tajc editing block
Plate 10.3. I:errograph lape recorder
1 0 .J
The Leecers-Rich !ape
Plate 1 o.5. Cerztral Recording Room H . 1 8
Plate 10.6. Engineer's control cubicle at the Paris studio in London, showing, in addition to
the ~wrmolcontrol baj. the three tape recordersfor local recording. The Studio Manager's
control cubicle can be seen through the far window
Plnte 10.8. Ficord I.-I recorder,
.vhot~.ittge . r t r e ~ ~small
w ~ size
Plate 10.9, Aragra
I B recorder
to know that a recording is in fact on the tape, because of the high
cost of repetition when expensive artists are involved.
The Erasing Process
It is necessary before new material is recorded on a tape for the
tape to be completely free from any previous magnetism. This is
achieved by passing it across the erase head. This consists of a
ring of laminations of magnetic material, having a small but finite
gap (Fig. 10.I), the laminations being magnetised by an alternating current passing through a coil wound round them. This
current at high frequency, as described above, will cause an alternating magnetic field to appear across the gap in the laminations.
This field will " bow " out into the tape with the result that as the
tape moves across the head each minute section of tape will pass
first an increasing alternating magnetic field which will take it up
to full magnetisation, and then immediately through a decreasing
magnetic field which will reduce this magnetisation to zero.
The Recording Process
The completely demagnetised .tape next passes across the recording head which is a similar ring of laminations, again having a
coil, through which, this time, the programme currents pass, having
been amplified by the recording amplifier. These currents again
produce a varying magnetic field across the gap in the laminations,
and this field magnetises the tape in sympathy with the programme.
In order for this to be achieved without distortion, it is necessary
to apply a b k to the recording head at the same time as the
programme. This is necessary because the relationship between
the magnetising force and the magnetism i n d d in a magnetic
1 5 ~
material, in this case the tape, is not linear-see Fig. 10.3. In the
absence of bias, a sine wave input would produce a recorded waveform wnsiderably distorted as sho\vm. Fortunately the magnetisation curve has portions which are substantially linear and the bias
assists in ensuring that the audio frequencies magnetise the tape
lo connect nvR-lincmily
on this linear part of its characteristic, as shown in Fig. 10.4.
In the early days of magnetic recording, a d.c. bias was used,
and whilst this was successful in reducing distortion it resulted in
a poor signal-to-noise ratio. The modem use of high frequency bias
reduces the distortion and ensures that the signal-to-noise ratio has
an acceptably high value.
Tbt Reproducing Process
The reproducing head is again similar in construction to the
recording head, and the changing magnetic state of the tape as it
passes the gap in the head induces a changing magnetic field in the
laminations which, in its turn, induces an alternating current in
the coil. Thii current is fed to the reproducing amplifier.
Equalisadon in Tape Reconllng
When a recorded tape is passed over a replay head, the voltage
produced a t the terminals of the head will be equal to the number
of turns on the core multiplied by the r& of change of flux in the
core. The rate of change will depend upon the speed of the tape,
and the fie- of the recorded material. If the recording has
been made at constant r.m.s. intensity, the output voltage from the
replay head will be doubled every time the recorded frequency is
doubled; that is, it will rise with frequency at the rate of 6dB/
octave (see Fig. 10.5). Clearly this is not a flat response, and some
equalisation is necessary. Equalisation for this rising characteristic is contained in the replay amplifier.
At very low frequencies, and at very high frequencies, further
difficulties occur.
At very low frequencies, due to the fact that the " magnets "
recorded on the tape are longer than the length of tape in contact
with the pole faces which contain the gap, not all the flw will
close round the core of the head, and some of the path of the flux
will be in the air. Because of this more losses occur, and the bass
response falls at a rate considerably in excess of 6dB/octave.
Fig. 10.5. Li'wqualrvd lap
rcmrding characlcnriic
Equalisation for this loss is usually only found in professional
recording equipment, and is situated in the recording amplifier.
At very high frequencies, as the recorded wavelength (in length of
tape) becomes comparable with the width of the replay head
gap, the output from the head will fall, until it reaches zero at the
extinction f r e q m q where these two are equal. At half this
I 60
frequency the output is 3dB down, Equalisation for this high
frequency loss is included in the replay amplifier.
I n addition to the above needs for equalisation, further losses
high frequencies, due to the fall in tape permeability and
the tendency for selfdemagnetisation to occur within the tape.
Again, hysteresis and eddy current losses cause still h t h e r high
frequency reduction. Equalisation for these is normally included
in the recording amplifier.
occur at
10.2.5. The T a p Transport Mechanism
As stated in the introduction, the function of this mechanism is
to ensure that the tape passes the heads at an absolutely constant
linear meed. This
nonnallv achieved bv means of a cabstan
driven a t a constant speed by means of an electric motor and antivibration coupling. The tape is held in contact with the capstan
by the pinch whet, a rubber- covered pulley held against the- tape
by spring tension. The driven tape speed may vary from machine
to machine, the standard speeds being 30, 15, 73, 3), 1it or even
in. per second. Of these, the first two and increasingly the
third are used for professional purposes, whilst the last four are at
the moment used in domestic equipment. The tape transport
mechanism is completed by two firther motors arranged to drive
the feed and takeup spools during recording and reproduction and
for the rewinding or spooling process. On professional equipment
and some of the better class domestic equipment, the spooling
speed can be varied; on most domestic machines, however, a
fixed speed is provided.
10.3.1. Advantages
(a) The medium-namely, the tape-may be used over and over
again. This is especially useful in a broadcasting organisation
where so much recorded material is required only for a short
(b) Recording and reproduction may be carried out by remote
control from the studio by means of relay-operated circuits.
(c) Editing, which involves cutting and re-joining, or the dubbing
of extracts to another tape, can produce a continuous taped
(d) Relatively long playing times are possible on each tape.
(e) Storage is fairly simple, although the length of store life is
(f) Many reproductions can be effected without loss in quality or
increase in background noise.
(a) Playback is not possible until the tape is rewound.
(b) A number of recorded passages are less easily superimposed
than with discs.
(c) Spurious magnetic fields may introduce noise to tape recordings
or partially erase them.
(d) The recorded modulation cannot be seen.
There are two methods of editing tape recordings. Firstly,
when an extra machine is available, the required passages can be
dubbed in sequence so as to build up the finished programme. This
method has the advantage that it obviates cutting of the original
tapes. Secondly, there is the cut-and-join method in which the
required passages are cut from the original tapes and a fresh composite tape built up from them. In either case the exact positions of
the tape from which the dubbing is extracted or the cuts made are
found by moving the tape manually across the replay head at slow
speed, listening to the tape to find the required gap. A mark is
then made on the tape in the position of the head gap and this is
the editing point. Tape editing by the cut-and-join method is not,
Fig. 10.6. Joining &pI
of course, possible when more than one track, each carrying a
different prog-e,
is recorded on the tape, without destroying
all the tracks except the one being edited.
When a tape is edited by cutting, it is essential that accurate cuts
be made so that different pieces of tape can easily be joined
together without dixontinuitics showing when the tape is subsequently reproduced. Some form of editing device is desirable, and
this may conveniently consist of a milled channel in a block of metal
in which the tape is laid, the edges of the channel being undercut
so that the tape is held firmly. A narrow slot at 45" to the tape
channel provides an accurate guide for a razor blade (see Plate
10.2). Joining of tapes may be by means of special adhesive tape
applied to the reverse side of the recording tape over a butt joint,
or a solvent may be used with a small overlap (Fig. 10.6). Normally
the adhesive tape joint is used for normal cut-and-join editing in the
studio or editing room, and the solvent used for " service joints "
when used tape is checked before subsequent re-issue. Great care
should be taken that no adhesive material gets on to the coated side
of the tape or it may foul the tape guides, causing hulty reproduction.
The FerrograpJa Tape Recorder (Plate 10.3)
This is a semi-portable high grade domestic machine and is
principally used by the BBC as a rehearsal recorder. The machine
is, however, used for transmission by a number of Colonial Broadcasting Stations. Two versions of the machine are available,
running at 15/74 and 74/39 in. per second respectively, and as
used by the BBC are provided with input and output circuits for
use with zero level programme at 600 ohms impedance. A second
input jack is provided for a low level input such as a microphone at
high impedance. Bass and treble tone controls are provided,
operative on playback only. The machine has no facilities for
monitoring the tape while recording.
Thc EM.L Magnedc Recorder TRIgo (Plate 10.2)
This is a highquality tape recording and reproducing machine
suitable for rack-mounting in a confined space, or for trolley-mounting. Tape speeds of 74 and 15 in. per second are provided, the
latter being the standard speed for BBC programme tapes giving
32 minutes playing time for 2,400 ft of tape. A number of studios
have been fitted with special sockets which permit operation of the
trolley-mounted TR/go, either for recording or for reproducing
to a channel on the desk or via an acoustic effects loudspeaker in
the studio.
A spooling control operates when the spool button has been
pressed, and gives continuous control over the tape speed, from full
speed forward to full speed reverse. Complete rewinding takes
1 ~ 3
under two minutes. The run and of controls are relay-operated,
so that remote-starting is possible. A timing indicafor gives a good
check on timings, and, being frictiondriven, may be set back to
zero as required.
Very similar in facilities and performance is the E.M.I. BTR/z,
the static tape machine employed in many recording channels.
zag.% Lccvers.Rich Tape Reproducer (Plate104)
This is a high-quality tape reproducing machine, normally
mounted as a transportable console, which can be operated in
studios using special sockets as in the trolley-mounted TR/go.
Two speeds are again provided, 74 and 15in. per second, but unlike
the TRIP, this machine is not relay operated fiom push-buttons.
The various functions are obtained by operating the appropriate
switches; a large switch on the right-hand side of the machine
controls spooling and normal playing, and in the spool position a
knob on the left-hand side gives continuous control over the tape
speed. Again unlike the TR/go, the change of reproducing speed
is accomplished by means of a switch with a knob on the left-hand
side of the deck, and a second knob on the panel on the front of the
machine must be turned to change over the equaliser in the reproducing amplifier. Both these switches have intermediate " off"
A monitoring amplifier is provided with its own gain control,
enabling pre-fade listening either on a small built-in loudspeaker
or on headphones. In the normal condition of the machine, the
starting time is much slower than the TR/go, but is is possible to
arrange that the capstan motor is running continuously, and in this
condition the operation is much speeded up. AU machines found
in studios have the capstan motor running fiom the time the
equipment is switched on.
The BTR/z, TR/go and Leevers-Rich machines are all capable
of remote control. Two types of such control are found.
In the first case, the remote facility may be used to start machines
in a studio cubicle from the studio desk, the machines having previously been set up, and switched to " remote
This facility is
useful when tape inserts have to be cued accurately into fast moving
programmes. Facilities for starting several machines in this way
may be provided.
The second type of remote operation involves the setting-up of a
central recording room in which recording staff supervise a number
of tape machines. Special circuitry at the studio enables the studio
manager to stop and start the tape, and to monitor the actual
recording. This means he is listening to the programme approximately one-fifth of a second late, which is a delay likely to cause
trouble on a fast-moving production, but is tolerable on talks, etc.
The P.P.M. reads the normal studio output.
10.6.1. The Broadcasting House Arrangement
In this, all talks and general purpose studios are connected to the
central recording room (Plate IO.~),and each is fitted with:
(a) Record off/mn key and red indicator lamp--used to stop and start
machines in the central room, which are set up to record.
(b) Monitorfeed-ring main point 10is connected to the appropriate
machine in the central room, both for record and playback.
Operation of the record off/mn key also switches on the studio red
lights, after a delay of 2 seconds, which has been introduced to
prevent the studio starting to speak before the machine has run up
to speed.
The arrangements at other centres, such as Bush House, differ in
detail from those described above, but the general principles of
operation are the same.
10.6.2. Reproduction
If remote controlled reproduction into studio pro&twm is required, extra circuitry is necessary to bring up the tape output
as a source in the studio. This may be brought to a special fader
mounted on a tape control unit, or as with the new Type B control
equipment at Bush House, the tape output is automatically switched
to the studio on Outside Source line I.
Reproduction of complete programmes into a given Domestic Service
is effected in the appropriate Continuity Suite.
An increasing number of studios are being equipped with local
recording/reproduction facilities. Being able to record, mix,
and edit a programme in step with the producer's arrangements for
rehearsal is useful on many types of programme, particularly since
it facilitates the techniques of recording sequence by sequence.
An excellent on-site recording" installation consists of three static
tape machines with linking console housed in a room adjoining and
communicating with the studio control cubicle. A sound-proof
window will
good liaison between the rooms, while permitting
extra editing or other work to be done as the studio rehearsal proceeds (Plate 10.6).
The use of miniature battery-operated tape recorders for on-thespot recordings, such as interviews, news reporting and sound
effects, has become very popular, and several machines of this type
are used.
The EM.1. Midget Recorder-Type L.2 (Plate 10.7)
This recorder is contained in a wooden rexine covered case,
14 x 7 x 8 in., and two versions are available. The earlier
version is valve operated and weighs 144 lb, including batteries. A
separate reproducing head and amplifier allows the recording to
be monitored on headphones if required. Playback is also possible,
but it is better to playback on static equipment and so conserve the
life of the batteries. No erase head is included and clean or preerased tapes must be used.
The record/replay switch is on the tape deck, and when this has
been put in the required position, and the drive roller latched into
position, the lid may be closed, and the machine started by means
of the battery onlof switch in the end compartment. Three
windows in the lid allow the level meter and the amounts of tape
cn the two spools to be observed. A geared rewind, operated by
hand, is also in the lid.
Nine Venner accumulators or dry cells are used to provide L.T.
and motor driving voltage; their life is about 6 hours. H.T. is
provided by two 67 a 5 volt dry batteris whose life is about 15 hours.
The newer version of this machine has a smaller transistorised
amplifier and the extra space which is thus available in the case has
permitted the inclusion of a monitoring amplifier and loud-speaker.
Input is usually from an S.T. & C. 4 0 y G moving-coil microphone
and the 4037 microphone can also be used. The 30 ohm impedance of these microphones matches directly the input circuit of
the recorder. The maximum recording time, with a 5 in. diameter
reel at 7+ in. per second, is 15 minutes using standard tape.
I 66
10.8.2. The Fieord Minktrac &coder-Madd rA (Platc 10.8)
This is an extremely small, transistorised light-weight machine,
its dimensions being 94 x 5 x 29 in., and its weight qf Ib. One
record/reproduce head is fitted and this means that no monitoring
from the tape can be done whilst the machine is recording. Unlike
the E.M.I. machine, however, the Ficord has an erase head and
so does not need to be used with pre-erased tape. The amplifier is
transistorised and a switch converts it &om record to play
function. A miniature loudspeaker operates on replay for monitoring purposes only. The machine is powered by six leadlacid 2 volt
miniature accumulators, each having a sealed plastic case so that
the battery is unspillable. A charger is provided having approximately the same dimensions as the recorder, and a fully charged
set of batteries will run the machine for two hours at 73 in. per
second. At the alternative speed of I $ in. per second this battery
life is extended to between three and three and a half hours, but
this speed is not used for broadcasting purposes.
The microphone used with this machine is a Grampian movingcoil and the playing time for a 300 ft spool of long-play tape is a
maximum of nine minutes.
The Naga Tape Recorder-Model xxrB (Plate I 0.9)
This is a miniature battery operated portable machine, made to
high quality professional standards. The dimensions of the
machine are 14i) x 94 x 49 in. approximately, and its weight is
I 5 lb I I oz including batteries.
The machine operates at three speeds, 15, 74 and 32 in. per
second, and a special feature of the design is the excellent speed
stability and freedom from wow and flutter. This is largely due
to a special electronic stabilising device.
Two inputs are provided, each having a separate fader so that
they can be mixed. One is normally for a moving-coil microphone
of 50-200 ohm hpedance, and the other a high level input at
higher impedance. An accessory unit can be obtained enabling a
capacitor microphone of the Neumann " KM " type to be used into
this high level input, power for the microphone being taken from
the recorder. Two outputs are provided, one a monitoring output
into headphones or loudspeaker, and the other an output to feed a
6 dB.
600 ohm line at a level of
The amplifiers are transistorised, and the machine is of the threehead type so that monitoring From the tape is possible during
Level indication is by means of a P.P.M.-type instrument so that
good peak indications are given.
The battery consists of twelve I -5 volt " U2 " type cells, and will
give an operating life of 10-12 hours depending on the type of use.
The batteq voltage is most critical when the machine is operated
at I 5 in. per second.
The machine will accept 7 in. reels of tape, giving 15 minutes
recording time at I j in. per second or 30 minutes at 74 in. per
second, using standard tape.
Recordings made on this machine can be comparable in every
way with those made on static professional equipment.
DISCrecords, as a means of storing sound, have become a major
source of domestic entertainment, and so naturally find their place in
a broadcasting o r g a h t i o n such as the BBC. hlany programmes
are built round commercially recorded discs, and such discs are
widely used as incidental music in dramatic and feature productions.
In addition to using commercial gramophone records-the
BBC has probably the largest gramophone library in the worldthe BBC pioneered the use of the Direct Cut Lacquer disc as a
means of recording programmes. Whilst this use of disc recording
has now been superseded by tape, a few specialised uses of the
medium still remain. For example, in the compilation of News
Bulletins, it may be desirable to use only a part of a particular news
despatch in a Gven transmission, whereas-in another transmission
the whole despatch, or a different part may be needed. The
decision as to exactly which extract is needed for a given bulletin
may have to be taken after the programme is on thk air, and so
some rapid means of selection is obviously necessary. Given
suitable equipment with discs this can be accomplished in a minimum of time. For similar reasons. sound effects for Features and
Drama productions have also bee; recorded on disc, and a very
flexible production routine has been developed over the years,
using a number of specially designed turntables running at 78 r.p.m.
Modem considerations of quality would dictate that both these
types of usage could be bettered by tape recording, but so f i r no
practical tape-reproducing equipment has been developed which
will pennit the rapid operation and flexibility necessary. In a
typical programme, as many as eighty or ninety effects may be
necessary, each of only a few seconds duration, and with disc
operation, given the specialised equipment, it is possible to find any
one of these, and play it into the programme in no more than five
seconds. Montages of a number of effects, each on a separate
disc, can be built up, and the composition of a montage can be
varied, if necessary, to take up any changes in the tempo of playing
a dramatic scene, for example bemeen rehearsal and transmission.
I 1.1.
In disc recording, the sound vibrations in the air are converted
into alternating currents by the microphone. These currents are
amplified and sent to the coil of the recording head, where they
give rise to an alternating magnetic field. This interacts with the
field of the permanent magnet in such a way as to give the armature a turning movement about the pivot rod. Thus, the cutting
Fig. r 1.1. Dirc rccmdfnghcud
stylus vibrates to and fro at the frequency of the original sound
waves, and will cut a lateral trace on the disc which is rotated at a
constant speed (Fig. I I . I ) .
It is necessary when recording to introduce bass cuts in the programme material in order to prevent the cutting stylus making such
a large movement that it tends to cut into the previous groove. It
is also desirable to introduce some high frequency lift in order to
overcome the worst effects of surface noise when playing the disc.
Some years ago both the bass cut and top lift were different in
amount from organisation to organisation and from recording
company to recording company, and so correct re-play conditions
for any given disc were hard to achieve. In the last fav years
international agreement has been reached on these recording charmtcristiw and standard playback equalisers are now fitted to almost
all disc-playing equipment (Fig. I I .I). Discs recorded before the
introduction of this standard will not of course be correctly reproduced but the standard has been chosen as a reasonable compromise
I 7O
and no serious error should result. U; in isolated caxs, extra
correction is necessary a variable firequency response control unit
should be used.
T d k spdddr are measured in revolutions per minute or r.p.m.
The speeds normally used are 78, 335, and 45 '.p.m. but recently
experimental discs and talking books have been recorded at 16
78 r.p.m. is still used for some commercial gramophone records,
and for BBC discs where ease of editing is a prime consideration.
Playing times of up to five minutes are possible per side.
335 r.p.m. is used for h e groove recording (microgroove recording) as found on commercial long-playing records. It has also been
Fig. r 1.2. Britizh Stanaiud/inc gmoa recording charocleristic
used for Transcription recordings with a coarse groove on a I 74 in.
diameter disc, enabling a 15 minute programme to be recorded
without a break.
45 r.jt.m. is also used for commercial long-playing records.
The number of grooves per inch on a disc is called the fitch. 104
grooves per inch are normally cut on BBC coarse-groove discs,
although I 20 grooves per inch are sometimes used to accommodate
long recordings of effects etc. The pitch on commercial 78 r.p.m.
records varies a great deal, but about 100-120 grooves per inch is
usual. Long-playing gramophone records have shallow grooves a t
about 220-350 to the inch. BBC fine-groove recordings have a
pitch of 240. In disc recording there is a limit to how closely the
grooves can be spaced, and how slowly the turntable can rotate, if
good quality is to be maintained.
Direct recording on a lacquer-coated aluminium disc has a
number of advantages and disadvantages:11.2.
r x.z.1.
(a) The disc is ready for immediate playback-no rewinding is
necessary as with tape.
(b) The rapid editing and fitting together of extracts is conveniently
done on discs (this is especially true of 78 r.p.m. discs).
(c) The same turntable desk will, with suitable equalisation, play
BBC discs and commercial gramophone records.
(d) Processing, so as to obtain records for permanent retention, or
many copies, may be accomplished by the same method as is
used in the gramophone industry.
(a) The recording is not permanent-in fact, very cadi11 handling
is necessary if a dozen noise-free playings are to be obtained.
(b) Greater skill and expertise are required to cut a good disc than
are necessary to make a good tape recording.
(c) Only a relatively short recording time is possible per disc.
(d) Loss of high frequencies tends to occur towards the centre of
the disc.
In order to process a direct recorded disc, a master or matrix
must be made, from which, by further processes, any number of
copies of the original recording can be produced.
The principal material for the making of pressings was previously
shellac, but more recently vinyl has been introduced. This gives
lighter and less brittle pressings, with less surface noise.
and storage of these new pressings calls for a great deal of care,
however, as they attract dust, and are marred by even tiny scratches.
In full-processing, the master is obtained by immersing the
prepared original in an electro-plating bath and " growing " a
negative of pure copper on to it. From this master, the mother is
obtained by another electro-plating process, and will have grooves
identical to those of the original. The stamper is prepared from the
mother, and makes a " positive " impression on the final record or
pressing, which is malleable at high temperatures, and becomes
hard when cooled (Fig. 1 I .3).
Half-processing is undertaken when only up to 50 pressing are
required. The master also becomes the stamper, after being
hardened by nickel and chromium plating. There is naturally a
A h : Fig. rr.3. SLrpI in @IpIuammng.
L 3 : F i g . r r . 4 Steps in
saving of time and expense, and many BBC discs are duplicated in
this way (Fig. 1 I .4).
This desk was developed 'some years ago for the playing of
BBC 78 r.p.m. disc recordings, as well as commercial gramophone
records. I t is being retained for as long as is necessary in accordance with production requirements. I t is a twin turntable desk,
and with it the playing of short extracts from any part of a 78
r.p.m. disc is possible with considerable accuracy. It will be
described in detail.
Reproducing Head Type E.M.X. 12 (Fig. I 1.5)
This is a medium-weight reproducing head fitted with a sapphiretipped stylus. A counter-balance gives a tracking weight of I +
The stylus fits into a tubular armature and is held through slots
by an elastic band and by magnetic attraction. It is important
that the needle should not be rotated, as a " flat " may have been
worn which will damage the disc. For this reason, and to prevent
11.5. E.M.I.
12 reproducing
unauthorised removal, the needle is sometimes cemented into the
armature tube. By modern standards, this pick-up is no longer
considered " light-weight ",and any future head is likely to be much
A two-position key has been fitted to meet the different requirements of BBC and commercial recordings. In the direct recordings
position, the pick-up is connected to the output circuit via an equaliser circuit which corrects for the BBC recording characteristic
which rises at about 4 dB per octave (Fig. I I .6). In the gramphone
78 r.p.m. recurding cAmuct2ktiGs
records andpasings position, fivther attenuation is introduced above
6,000 c/s to reduce surface noise, and the overall level is brought
down 4 dB to correspond to the level recorded on BBC discs (see
Fig. I 1.7). A step-by-step attenuator (& dB in 2 d B steps) is
- 140.1
Fig. rr.7.
Efki of +&on
k q on TD!7
provided inside the desk, and is adjusted by the engineering staff
to compensate for the varying sensitivities of different heads.
Pre-fade Lietening
As may be seen from the block diagram (Fig. I I .8), a separate
output is taken to each pre-fade amplifier. A separate tsvo-position
key for each turntable switches the headphone circuit either to the
pre-fade amplifier or to the programme selector switch.
When the turntable is rotating at its correct speed, the beams of
light from the neon lamp appear to be stationary. This is because
the number of holes round the turntable has been calculated so that
each hole is exactly replaced by the next, in step with the alternating brightness of the ax.-driven neon lamp. Now, with 50 c/s
mains, the neon glows bright loo times per second, or 6,000 times
per minute. If, therefore, 6,000 holes pass the neon per minute, the
pattern will appear stationary, and for 78 revolutions per minute
there must be 6,ooo/78-i.e.
77 (approximately) holes equally
spaced round the edge of the turntable.
For the same reason, 77 green bars are printed round the edge
of the labels on BBC discs, giving a useful stroboscope indication
when looked at in light from 50 cis mains.
I I.I.
Turntable desk TD!7 showing D E T U i r
Plafe 11.2. BBC Jinr-groocr
reproduring d d DRD:5
However, if the mains frequency is not exactly 50 c/s at the time
of reproduction, perhaps due to a power-cut, setting by the stroboscope will give wrong speed adjustment. It is necessary, theh, to
play a record of I ,000 c/s tone on each turntable in turn, and adjust
the speed until the pitch lines up with tone sent by the control room
(for perfect speed adjustment beats between the two tones should be
eliminated). Recording staff are instructed to record a short band
of tone at the beginning of at least the first two discs of a programme,
and this tone should be synchronised with the control room tone a
short time before transmission.
x x 4.5.
Groove-LoCPting Unit GLUjgB
Parallel tracking is used, which simplifies the design of an accurate
groove-locating mechanism, but has so far prevented the introduction of an upto-date tight-weight pick-up. The pick-up head
is attached to a light-weight tracking-arm which is held by a carriage.
This runs on a polished carriage-rail, and its six ball-races ensure
minimum friction and sideplay. The rail itself is supported in
ball-races, within anti-vibration mountings.
Operating the lifting lever causes a bar to press down the counterbalance, so raising the pick-up. A drum at the end of the carriagerail gives fine adjustment of the pick-up position. A scale with
Fig. I 1.8. TD,7--cirrw'1 diagram
1 7 ~
divisions, each representing ten grooves at a pitch of 104 grooves
per inch, is fixed to the cover of the GLU, and this helps to locate
narrow bands on the disc.
The simplest procedure for groove-locating is described in this
quotation from the BBC Recording Training Manual, which
should be consulted for a more detailed description of the apparatus:
" The disc is played until the beginning of the required excerpt
is heard on headphones connected to the pre-fide amplifier;
the pick-up is then raised by the lever, the turntable continuing to
rotate; the h e control is then turned so as to set the pick-up
back by a small amount. The precise setting of the h e control
is a matter of practice rather than skill and the accuracy obtained
is sufficiently high for most requirements. A little thought will
show that the application of this system to 334 '.p.m. d
m would
be less satisfactory, because of the increased programme time
occupied by a single groove ".
As is well-known, the best accuracy obtainable, using the simple
procedure described above, is plus or minus half-a-groove. To
improve on this, it is necessary to halt the pick-up in a particular
part of the required groove. Doing this by hand necessitates
braking the turntable suddenly and spininng it on cue. Unless
this is done very gently, the sapphire digs in and puts a click on the
disc. If it is done roughly, damage is possible to the turntable
motor (not to mention the disc), and in particular a great strain
is put on the anti-vibration coupling. The motor driving-shaft is
coupled to the turntable spindle via two plastic discs which are
riveted together. This virtually isolates the pick-up from motor
vibrations, but, of course, makes gentle handling necessary.
11.4.6. Qoick-start Devices
The necessity for accurate cueing of disc inserts, and the everincreasing Speed of operation has necessitated the introduction of a
quick-start device for 78 r.p.m. turntables.
In the original device, the operation was by means of two pushbuttons, one red, one green.
Pressing the red button causes a pair of rubber-covered blades to
press against the underside of the disc, thus lifting it just clear of the
turntable, and bringing it to rest almost instantaneously. The
pick-up is still resting on the disc, and the setting-up procedure is
simply one of arranging that the disc is stopped about half-arevolution before the desired cue.
Pressing the green button causes the lifting blades to firll away,
and the disc to drop back on to the already revolving turntable. As
soon as the relatively light disc touches the rubber friction mat of
the turntable, it starts to rotate and quickly reaches the correct
speed. With practice it is possible to fade up without any trace of
L L WOW ", and start the disc at any required cue.
The mechanical action is simple, and may be appreciated by
reference to the rough diagram in Fig. I 1.9. Pressing the red button causes the pivoted lever to raise the Lifting blades until the lever is
I I .g.
Protorypr quick-start dmia
engaged by the catch. This holds the disc stationary and clear of
the turntable. It is possible to " edge " the disc round by pushing
the red button just short of the " engaged " position.
Pressing the green button disengages the catch, so that the lifting
mechanism and disc return to their lower position, and the disc is
accelerated auicklv to normal s~eed.
An impro;ed {uick-start divice has been produced by BBC
Designs Department, and given the code name DETU/I-" Dropstart Editing Turntable Unit
The red &d green buttons are replaced by a single lever, and the
two lifting pads are replaced by a three-point lift. The original
12-in. turntable is replaced by one of g in. diameter, which permits
a clearance for the lifting mechanism for 10 in. discs. A paper
strip round the edge of the turntable carries the stroboscope markings, and is illuminated by a pair of neon lamps. A new speedcontrol knob has been fitted--see bottom left-hand corner of
photograph (Plate I I. I).
This two-speed desk is used by the BBC mainly for the reproduction of discs recorded at 333 r.p.m. The development of tape
1 7 ~
recording for many programmes has considerably rcduad h e usc
of" slow-speed " discs. A rubber tym is fitted round the rim ofthe
turntable and the driving roller, which prnscs against this, has two
diameters. The smaller of these drives the turntable a t 33) r.p.m.,
and the larger at 78 r.p.m. The appropriate speed is obtained by
operating a speed-selector lever (Fig. I I. I 0).
After use, the locking-nut must be loosened, and the lever moved
to the left to remove the drive pressure. Failure to take this precaution may result in flats being impressed on the turntable tym,
with consequent " wow " and rumble.
This desk is normally used to reproduce 17) in. slow-speed
transcription recordings, having a coarse groove, although some
desks have been modified with an additional pick-up for fine-groove
long-playing discs.
This desk is a specially designed huntable and pick-up for
reproducing finegroove records at speeds of 333 and 45 r.p.m.
A lightweight crystal pick-up is provided and them is an optical
system associated with the pivot of the pick-up arm which gives
an approximate indication of the position of the pick-up head on a
ground glass scale at the back of the instrument. The accuracy is
such that resetting to a given division on the scale ensum that the
pick-up is within five grooves of the passage on the disc.
A sensitive quick-start device is provided; in this case the turntable rises to meet the disc rather than the dix dropping, as in the
case of the TD/7. The speed of starting is such that the disc is
running accurately to speed within one-quarter of a revolution. An
improved stroboscope is provided consisting ofa large plate revolving
on the turntable having the stroboscope markings provided by
means of holes drilled round the edge : these are viewed by means of
a neon lamp which has a special pulse sharpening circuit to enable
a clear indication to be obtained. Indicating lamps are provided
to show whether 45 r.p.m. or 33) r.p.m. has been selected and a
red indicator panel is illuminated when the fader is turned up.
Pre-fade listening on headphones is available, and also a top cut
filter providing a tail-off above ro kc/s, which is usell in the removal of distortion which is sometimes present on worn discs.
The R P ~ / is
I a high quality desk for the playing of all types of
disc with the exception of 16 in. slow-speed coarse-groove types.
Hence it will play all 78 r.p.m., 45 r.p.m., and 334 r.p.m. coarse or
fine-groove records.
In contrast to the DRD/5, the RP2/r has two turntable and pickup units mounted in the same console cabinet. Each unit has its
own equaliser pre-amplifier, powered from a common power supply.
The quick-start system of the DRD/5 has been retained, but is now
electric motor operated, and, in consequence, has a slight time
delay. The optical groove indication is also retained, and there
is a simplified raise/lower mechanism for the pick-up arm.
Pre-fade fkilities are provided, and a pre-fade gain control has
been added giving a boost of some I o dB if required.
The pick-up head is a magnetic variable reluctance type of the
turn-over variety, allowing for change of stylus for coarse and finegroove discs. The equalisation for the two types of disc is changed
automatically with the head by means of a miniature micro switch
in the head itrelf, which operates a relay on the amplifier units.
A aon-linkage with the turntable speed switch causes the speed
indicator lamps to flash if the stylus selected is unsuitable to the
disc speed, but this d o g not prevent the disc k g played in this
condition Care must, therefore, be taken to ensure that finegroove disa are not played with the stylus for coarse grooves, or
damage to the disc will result.
The description which follows is concerned with the Broadcasting
Chain as developed by the BBC, and is especially suited to the
type of sound broadcasting found in the United Kingdom where one
basic organisation, together with regional stations, supplies programmes covering the whole country. Other organisations may
find different systems more suitable to their requirements, but many
of the individual stages in the process described below will be of
The sequence of events through which a BBC programme passes
on its long journey from the studio to the listener's home is as
follows (Fig. I 2. I).:(a) The sounds in the Studio are picked up by the microphone, and
converted into an equivalent low level electric current.
(b) This is mixed with other sources where necessary in the Control
Cubicle, amplified and passed to continuity.
(c) In the Contimn'~Suitc the output of the studio is mixed with
other sources or interspersed with continuity material, and
passed to the control room.
(d) In the Control Room the programme is suitably amplified and
passed to the appropriate land line.
(e) The Post O#ce Lines carry the programme to one of more
transmitting stations.
(f) At the Trunsmitting Station, the programme is amplified, combined with radio frequency carrier currents (the process known
I 80
I 2. I .
Light Progmmme continuily suite
I 2.2.
Continuily desk
Above: Plate 12.3. The control room. Below: Plate 1 2 . j . Pontop Pike transmitting
station during conrtruction, showing the tubular seclion of the main TV mart for radiafing
VHF sound transmissions
as modulation) and fed to the transmitting aerial which radiates
the combined signal.
(g) The signal is picked up by the receiving aerial and passed into
the re~eiverwhere the programme current is removed from the
carrier (the process known as demodulation or detection),
amplified, and fed to the loudspeaker which vibrates at the same
frequencies as the original sound in the studio.
The studio manager's responsibility for the programme may be
taken to end at the point where it leaves the studio. I t is felt,
73.e pmgrmnrm chain
however, that a brief description of the functions and difficulties
of subsequent links in the chain will provide useful extra knowledge.
Following this analysis of the broadcasting chain there will be a
discussion of its imperfections as a means of producing faithful
reproduction of programmes in the listeners' homes.
The continuity engineer is responsible for technical monitoring
of the programme as a whole, and initiating engineering action in
case of fault. It is also his dutv to select subseauent source circuits
and carry out routine tests on each, prior to transmission. Finally,
he exercises control on the volume of fill-ups and other contributions
from the continuitv studio. Studio and outside broadcast items are.
of course, controlled at source, and no subsequent control is normally exercised by the continuity engineer (Plate 12.1).
Continuity working was introduced in the BBC in 1942, and the
layout of the control equipment has been modified only slightly
since that time. The various sources-studios, reproducing rooms,
etc.-appear on a series of source jacks. These are selected one
by one as the day progresses, and plugged to one of the input jacks
H I G H - Q U A L I T YS O U N D
assodated with a four-rhannrl mixer. The output of the mixer
is c o ~ e c t e dto an amplifier, which feeds the main programme to
the control room.
The desk also incorporates pre-We and programme meter Mties which are used during the prc-transmision tests on each source
(Fi. 12.2).
The equipment in the continuity studio adjoining the continuity
cubicle includes a microphone and at least two gramophone turntables complete with fader controls. There is also a main fada
El I :
which allows the announcer to fade down the main programme to
superimpose announcements, or to M e it out completely because
of technical faults or unsuitable programme material. The
announcer normally listens to the main programme continuously
on a loudspeaker which is automatically silenced when the microphone is live. He can switch the loudspeaker or headphones to
the faulty contribution, while filling up, to check when conditions
are favourable to rejoin it (Plate 12.2).
CONTROL ROOM (Plate 12.3.)
The control room engineers receive programme contributions
from a variety of sources, and d k t them to the appropriate
destinations. This involves a host of incidental operations, including the setting up and testing of routes according to the programme
schedule, monitoring miscellaneous programmes, and keeping a
check on the various circuits in order to trace and correct ~ H u t Q
with the least possible delay.
The control room is also a kind of switchboard for communications and telephone connections between all points h m programme sources to destinations. The staff of the control room also
carry out the day-today testing of all studio equipment, as well as
the equipment in the control room itself, and deal with any faults
and technical problems which arise in rehearsals, etc.
The network of programme lines interconnecting the various
studio centres and transmitting stations of the BBC form what is
called the Simultaneous Broadcast (S.B.) System (Fig. 12.3).
These circuits are more or less permanent, and are supplemented
on a short-term rental by special circuits to meet Outside Broadcast and other commitments. The routing of contributions into
the Light Programme, for example, is continually changing throughout the day, whereas the outward connections to the Light Programme transmitters remain fixed. The focal point in this network is the Light Rogramme continuity suite in Broadcasting
H o w (see map).
A contribution fiom outside London, say from Belfast, is routed
via intermediate BBC stations on S.B. " links ", as they are
called. Similarly, the output from Continuity in London travels
back to the Northern Ireland Light Programme transmitter at
ri.pnaParvey. There are nine links in this complete chain.
The intermediate stations do not simply pass the programme on
as it is received. Los in volume takes place along the line, and to
prtsave a satidkctory ratio of programme to noise (better than
40 dB) amplification is .n
A furtha complication arises
from the l3ct that the line loss is not the same at all fiequencies.
In fh,
each d o n of line has a characteristic response, depending
on its lengh, tempnature, etc., which may be different from any
other. The only characteristic which most lints have in common
is that loss tends to be greater at high fiequencies than low. In
addition to the ampyicr inserted at intermediate and receiving
stations, it is therefore ntcessary to include an cquulisr.
An equalisa is an electrid circuit made up from resistors,
capacitors and inductors, designed to give as near as possible the
mirror image of the response of the line. The effect will be to restore
the original quality More passing the programme into the next
section of line. i\'lien the distance between BBC stations is more
Fig. X2.3. Distribution nctwo~ksfor BBC Light Programme
than a few miles, it becomes the responsibility of the G.P.O.to
carry out such amplification and equalisation at intervals as may be
necessary. The stations set up for this purpose are known as
On amval at the transmitting station, the programme is suitably
amplified, equalised, and fed into the transmitter itself. As might
be expected, the transmitter valves and other components are
physically much larger than their counterparts in studio equipment, since the power developed by the transmitter may run into
hundredsof kilowatts-compared with zero level, the normal sending
level in studio centres, which is o-OOIwatts. I t follows that a
number of special problems are met in transmitters, such as insulation against the very high voltages used, provision of water cooling
systems to dissipate the enormous heat generated by the large valves,
etc.-but there is not space to describe these in detail. Instead,
two aspects of transmission only will be discussed.
Propagation characteristics on the different wavebands.
The relative merits of amplitude and fkequency modulation.
Pmpq&onofRadio Waves
We have seen in Chapter q that a wire carrying current is sur-
rounded by a magnetic field. Further, passing an alternating
current through the primary winding of a transformer results in an
e.m.f. being set up in the secondary. If this kind of" action at a
distance" could be extended over miles, and even thousands of
miles, world-wide transmission of signals would be possible.
Unfortunately, the problem is not as simple as the above remarks
might suggest. Early experimenters in wireless transmission found
that the radiation fiom an aerial carrying current a t studio frequencies (a few thousand cycles per second at most) was very weak
indeed. I t is only when current alternates at what are now called
radio fiequmics-15 kc/s up to about 3,000,000 Mc/s or morethat energy is thrown off from the aerial in such a way as to travel
over long distances.
Radio waves possess electric as well as magnetic properties, and
are therefore called e l c c i r ~ f warrcs.
They travel in free space
at a velocity of 3 0 3 , m , m metres per second (about 186,000 miles
per second), which is the speed of light. In fact, they are similar
to light waves in every way, except that they are of much longer
I 86
wavelength. T o calculate the wavelength of a radio wave whose
ikquency is known, we have recourse to the formula which we
used earlier for sound waves:-
h = - or wavelength =
where frequency is in c/s and wavelength in metres
The wavebands in use for broadcasting have been divided for
convenience as shown in Table I 2. I .
The propagation characteristics are found to vary considerably
for the different wave bands (Fig. 12.4).
The ground wwe is the name given to the direct ray which leaves
the aerial and travels along the surface of the earth. The distance
over which the ground wave will carry effectively is found to Eall
Fig. 12.4. Propagation diagram 5h+
sky -; (c) direct urrmc
from some thousands of miles on the Long waveband to about xoo
miles on the Medium waveband, dependent upon the power
radiated. Further reduction in wavelength continues to bring
down the range of the ground wave, until at V.H.F. the ground
wave is virtually non-existent and reception is only possible by the
direct wave at points which have a " line of sight " view of the
transmitter. This is usually a maximum of about 50 miles, for
normal transmitting and receiving aerial heights, depending on the
power radiated, and also on whether there are obstructions such
as hills or large buildings in the path of the transmission.
The sky wave is the name given to the indirect ray which leaves
the earth's surface at an angle, and only returns after reflection from
Table ma.
Long waves
Medium waves
Short waves
above 1,ooo metrea
200-500 meI+IW metres
7'3-4'4 In3-49rmaYs
I -7-1 -4 metres
045-0.5 metres approx.
0.5-0-3 meapprox.
15048 kc/s
525-1, 5 kc/s
8 bands in range 6-26
414% Mc/s
the layers of ionised gas (the ionosphere) which exist at heights of
about 60-100 miles above the earth.
Little use is made of the sky wave for transmissions on the Long
waveband. On medium waves, use is sometimes made of the sky
wave but it is something of an embarrassment since it may be
returned to earth at points where the ground wave is still fairly
strong. The different distances over which the waves have travelled
may cause them to arrive out of phase, so there is a region in which
fading takes place due to interference between the two signals. Reflection of medium and long waves by the ionosphere only occurs
after dark, so this interference is more likely to occur at night.
It is on the Short waveband that the sky wave is used extensively,
and the radiation is purposely angled to the horizontal to achieve
a skip distance appropriate to the required destination. It is not
proposed to go into detail on the many problems of short-wave
Woac brmd
Long waves
Medium waves
Short waves
V.H.F. band I
V.H.F. band I1
V.H.F. band I11
U.H.F. band IV
U.H.F. band V
Microwave band
[email protected] of Wave
ficiipnl Uses
mainly ground wave
ground wave and sky wave
long range broadcasting
short range broadcasting
long range broadcasting
(at night only)
world-wide broadcasting
television-channels 1-5
f.m. sound broadcasting
television--channels 6-13
yet to be developed for
yet to be developed
; sky wave
direct wave
I direct wave
; direct wave
direct wave
direct wave
' direct wave
propagation, except to say that for a given sending angle, t h a e is a
limiting frequency above which waves penetrate the Layers of the
ionosphere and are not returned to earth. Again, the effactive
height of the ionised layers is continually varying with seasonal and
hour-to-hour changes in the sun's position. The choice of a
suitable wavelength for transmission to New Zealand, for example,
will be extremely complicated, since of the several hops (reflections)
between the earth and the ionosphere which are necessary, if the
first takes place at midday in midsummer, the last will take place
at nearly midnight in midwinter.
At very high &equencies, the ground wave is very quickly absorbed,
and no predictable reflection of the sky wave takes place h m the
ionosphere. Reliable transmission is only possible, therefore,
over optical distances with no obstructions.
The above information is
'sed in Table 12.2.
So tkr we have been concerned with the transmission of the radio
freauencv wave. But the actual information that is to be broadcast
is contained in the programme currents at audio frequencies. It is
now necessary to examine the methods used to impress this low
frequency information onto the radio frequency " &ex
" signalthe process earlier called modulation. There are two systems of
Fig, r2.g. A m p l i t d modulation
modulation in common use, amplitude modulation and frequency
(a) In mpBu& modulation (a.m.), the amplitude of the carrier
current is made to vary in accordance with the amplitude of the
programme current, the number of variations per second being
equal to the programme frequency. The process is shown diagrammatically in Fig. 12.5, where the audio signal is seen to
appear as the emelope of the carrier wave.
The degree of modulation is the ratio of the audio and radio
amplitudes expressed as a percentage. For example, if the a.f.
amplitude is half the r.f. amplitude, we have 50% modulation.
Fig. 12.6. Frqrmgmod&wn
The maximum possible degree of modulation is 100% which
results in the combined amplitude rising to twice the r.f. amplitude and falling to zero in alternate half-cycles. Amplitude
modulation is used in television broadcasting, and until recently
was used for aIl sound broadcasting in this country.
(b) In frequm~ymodulation (Em.), the frequetlcy of the carrier current
is made to vary in accordance with the programme amplitude,
the process taking place as in a.m. at the programme frequency.
In BBC Em. sound transmissions a frequency swing of f 75 kc/s
is taken to correspond to maximum modulation (Fig. 12.6).
(Compare 100%modulation in a.m.)
It follow that f.m. transmissions are invariably placed in the
V.H.F. bands where the swing to either side of the carrier frequency
can take place without crowding by adjacent stations. The
V.H.F., Em. sound transmissions of the Home, Light and Third
programmes are now available to practically the whole of the
population of the United Kingdom.
In addition to the freedom from interference by adjacent stations
achieved in Em. transmissions, there is the important advantage
that electrical interference from refrigerators, hair driers, etc. (a
q e a t nuisance on a.m. reception in large towns) is much reduced,
due to the fgct that the Em. recciva is insmsitivc to the mainly
amplitude modulation of this type of interfercnce.
V.H.F., Em. transmission and reception
enables a higher
quality of sound reproduction to be achieved.
A detailed description of receiver circuits is beyond the scope of
this handbook.
It is impossible to achieve completely Gthfd transrmsslon or
recording of programmes. If ideal reproduction of the sounds was
possible, the listeners would receive the sounds with exactly the
same degree of realism as if they were present in the studio. Fortunately, it is possible to obtain adequate reproduction within the
limits of the system for most purposes.
Consider the following list of ways in which reproduced programmes can depart from true fidelity:
(a) The l o u i r p e h size is h e d , and yet it takes the place of all programme sources from orchestras to news readers.
(b) The mrmoplronic systcnr of broadcasting (a single channel feeding
a single loudspeaker) deprives us of the ability to locate the
directions of sounds.
(c) The l o u d r p d h volume will usually be different &om that of the
(d) The acoustics o
f the lishing room will be added to that of the
(e) The d v i c rmrgc is compressed to less than 30 dB.
(f) The nrirrophcmc balatue is subject to the taste and e x p u i ~ 1 c eof
the individual studio manager or engineer.
(g) Noise is generated in the programme chain.
(h) Non-linear d i h d o n tends to be generated in the programme
In this list, which may not be exhaustive, (a), (c) and (d) are
seen to be beyond the control of the originators of the broadcasts
or recordings. Methods of reducing the harmful effects of (e)
will be discussed in Chapter 14, and (f) is largely a matter of careful
selection and training of balance personnel.
Something can be done by engineering and S.M.staff to minimise
the effects of (g) noise, and (h) distortion, and the varieties of
these which can be met will now be briefly summarised.
12.7.1. Types of Noise
(a) Acourtic nobe in the studio originates from a number of causes
and is reduced by carehl sound-proofing, carpeting, etc.; it is
always high in television studios, which is one of the reasons for
using cardioid microphones.
(b) Amplifier noire takes the form of a steady hiss. It is caused by
the impinging of the electron stream on the anodes of valves,
and by the random movements of electrons in the amplifier components.
(c) Lowfiequenc- irum is associated with the a.c. mains supply. I t
is at the mains frequency-50 cis-and possibly harmonics, and
results from faulty rectification of the supply, faulty screening, or
placing microphones, etc., too near electrical apparatus.
(d) Line noise is of many different kinds. Examples are cross-talk
due to pick-up from other Post Office circuits, and random
thermal movements of electrons. The latter effect is present
even in short wires.
(e) Rcuioer noise in the listener's set includes interference h m
other stations, static from electrical apparatus, and usually some
mains hum. Listeners to the V.H.F. Em. transmissions experience
much less noise from machinery etc., and are very seldom troubled
by interference from other stations.
It is important for studio managers and engineers to remember
that they monitor programmes at a point of low noise level, compared with listeners, and this should be taken into account when
setting the dynamic range and the relative level of sound effects, etc.
1 2 . 7 ~ TgpeeofDbtorha
In general terms,distortion is said to be present when any change
in waveform takes place between two points in a transmission
system. Of the various forms of distortion, attenuation distortion
is dixussed here at some length, while only definitions are given
for other types.
(a) P b &&dun is present when the transmission time through
the system is different for different frequencies. It is not as a rule
serious at audio frequencies, except on very long line circuits where
it may distort transient sounds in speech and percussive music.
(b) flon-linear distortion is present when the properties of the
system vary for different levels of programme input. I t ~ ~bef l
introduced by circuit components which have response characteristics which change with input amp1itude-e.g. an amplifier when
overloaded or a transmitter when over-modulated. Son-linear
distortion may be subdivided into the following varieties, all of
which are not necessarily produced together:Amplitude distortion-variation of gain with input volume such
as may occur in a limiter at the instant of overload.
Harmonic distortion-production in the output of harmonics of
the input frequencies.
Intermodulation distortion-production in the output of sum and
difference tones of the input frequencies.
The effect of the last two forms of distortion is to produce a harsh
quality, and to distort the timbre of musical instruments.
(c) Atfmudion distortion (sometimes called ficquenq dirtortion) is
present when the properties of the system vary with frequency. It
Fig. 12.7.
Tmfmofal&miatim disrmiton-lop
cut and bass cut
may be introduced at any point in the programme chain--studio,
microphone, amplifiers, lines, transmitter, receiver or loudspeaker
-where the response to frequencies from 50 c/s to at least ~o,ooo
c/s is not uniform.
Loss of high frequencies (top cut) reduces the intelligibiiity of
speech and the " brightness " and " attack" of music. Loss of
low frequencies (bass cut) produces a thin quality and lack of weight.
When only a narrow band of frequencies is transmitted-both top
and bass cuts present-a telephone-like quality results. Boosting
of certain frequencies also causes a departure from true quality
(Fig. 12.7).
Provided attenuation distortion at a given point in the chain is not
too acute, it is possible to pass the programme through an equalising
circuit (see Section 12.4) and restore the original quality.
THISchapter summarises standard practices in the positioning of
microphones. It is intended to help newcomers, and form a basis
of experiment for more experienced studio managers. These notes
are based on the experience of many people in balancing programmes, but cannot, of course, take the place of actual experience.
Where there are drawings or photographs of layouts it is essential
to regard these only as examples chosen from the many possible
solutions to the given problem--e.g. the final position chosen for a
piano microphone will depend on the studio, the player, the
particular piece of music, and even the particular piano, and cannot be rigidly laid down.
Two-sided ribbon microphones are considered throughout, except
where stated. Other types of microphone, cardioid and omnidirectional may be used under suitable circumstances.
Before considering some of the particular cases that will arise in
the positioning of microphones it is necessary first to consider in
general the effect to be expected when more than one microphone
is used at once.
The use of more than one microphone can give rise to undesirable
effects, since the microphones will tend to interfere with each other.
1 ~ x . x . Spedal Case of Two Microphones Equidistant from Source
If a sound source is placed equidistant from two microphones, and
identical terminal connections are made to two channels of a
mixer, the resultant currents in the mixer will be in phase-i.e.
yill be " in the same sense
It follows that they will add at the
input to the amplifier (see Fig. I 3. I (a)).
If one of the microphones is now reversed (see Fig. I 3.1 (b)), or
terminals I and 2 interchanged, the diaphragm movements will
Fig. 13.1. Inrnfnmcc effect of
turn minophones cquidkht Jkm sour-(a)
in prUrre,
be in step, as before, but currents of opposite sense will result due
to the new positions of the poles of the magnet. In this case, the
outputs of the two microphones tend to canccl at the input to the
amplifier-the microphones are said to be " out of phase
laboratory conditions, and when the microphones have identical
response, the cancellation may be complete.
I n practice, of course, such a positioning would be avoided, but
whenever a layout approximating to this has been made necessary
Fig. 13.2. Gnnal cosc of tum microphones at &&&
d b
tanas frmn sanuce (a) b d y blacrd, ( b ) mcmrrcl angling fw
miRimMIpicrt-upon Mic. I
in platform discussions-a
out to h d the better position.
reversing test should be carried
If we now cansidtr the case when different distances apply (Fig.
13.2), we see that the diaphragm movements are no longer in
1 9 ~
phase, since the sound wave reaches the M e r microphone
(Mic. I ) slightly later. But there is one frequency (and harmonics)
for which the extra distance corresponds to a wavelength and for
which addition of the electrical outputs will take place. Similarly,
another frequency (and harmonics) exists for which this extra distance is half-a-wavelength and cancellation will result.
This partial cancellation and addition of different frequencies
amounts to distortion. and since s~ecialdistances have n i t been
considered, it follows that some distortion always results when two or
more microphones (of any type) are faded up together.
In practice, it is usually possible, with all but omni-directional
microphones, to reduce the distortion below the level at which it
becomes serious by suitably angling the microphones so that no two
r n i r r o p k @k up any source at anything like equal strength-g. in
Fig. 13.2(b), the source is placed effectively in the dead side of
Microphone I , and the contribution of Microphone I to the mixed
output may be ignored (so far as source S is concerned). Using a
close technique, as in multi-microphone dance band balances,
will further reduce this trouble.
Here we have the simplest type of balance problem, with one
microphone placed centrally in front of a single speaker. The aim
in a broadcast talk is to obtain faithful, pleasant transmission of
Fig. 13.3. Sbndmd position
fm broadcat talk
the voice and to make it sound as it would in ordinary living-room
acoustics. Given a good studio, the greatest single factor affecting
the balance is distance from the microphone.
I t is very important to avoid working too close, with the resultant
exaggeration of bass fi-equencies. A minimum of 2 ft should be
used, or 18 in. with a " corrected " microphone point (see Chapter
5). If a close technique is ever used for some special effect-such
" Mystery Voice " in " Twenty Questions "-it may be
necessary to warn the speaker that distortion is likely on consonants
like " "
The trouble may be partially reduced if
he speaks at a slight angle to the microphone.
Perhaps the most common acoustic defect in small studios is
boominess due to resonances. It is often possible to reduce the
distortion due to these resonances, and their accompanying pattern
as the
Fig. 13.4. Use of m3t-rest to correct looking down on to snipt
of standing waves, by exploring the available space. Working
towards a corner may be tried, or placing the speaker near the
centre of the studio, with the microphone slightly off-centre.
As fsr as possible, the speaker should be made comfortable, and
allowed to take up any position at the table he likes (Fig. 13.3).
However, positive action must be taken to discourage faults like
speaking down on to the script, which causes distortion due to the
interference of direct and reflected sounds, and the fact that more
Fig. 13.5. T&g of AXB
high frequencies are reflected than low. A script-rest will help,
as shown in Fig. 13.4 I t is to prevent reflection of sounds into the
microphone that studio tables are made with tops that are transparent to sound. Holding the suipt in front of the microphone is
another common fault, and results in selective masking of high
frequencies, and further distortion.
When a speaker is noticeably sibilant, some improvement may
result with a ribbon microphone if it is tilted as shown in Fig. 13.5.
This will be more.effective with older types of ribbon microphone
having fairly large dimensions. In this case, attenuation of high
frequencies will result since the live angle in the vertical plane is
narrower for high frequencies (see Chapter 5). This tilting must
be applied with discrimination to improve the overall pleasantness
of the reproduction, without reducing intelligibility or changing
the character of the voice.
Its double-sided nature makes the ribbon microphone very useful
in studio interviews and discussions with up to four speakus. The
AXB type ribbon is sometimes thought because of its size, to interfere with friendly discussion, and a dropped tray on elastic suspensions is let into some talks tables to allow the speakers to see
each other more easily (Fig. 13.6).
When more than about four speakers are involved, a single
ribbon microphone is often not sufficient. As Fig. 13.7 shows, the
Fig. 13.6. Minophanc let into rwllofl of talks &able
live angle of 100" is too narrow to accommodate three heads at the
normal working distance, and a greater distance may be impractical
because of the studio acoustics. I t is possible to get over this
dificulty by using two ribbon microphones, fading them up
separately when a script is available. But the use of both microphones faded up together (unscripted programmes) requires careful
handling, for the reasons explained in Section I 3.1.
At the present time a cardioid microphone is usually placed with
its working axis vertical below eye-level to give equal pick-up of all
speakers (or hcing downwards from above eye-level). The Neumann KMM, A.K.G. C I 2 and C 28 and the S.T. & C. 4033 microphones have all been used in this way with good results (see Plate
The above remarks apply to " round table " discussions. If the
for platform discussionsspeakers are placed " in line "-as
suitable angling of microphones (to avoid picking up any speaker
on more than one microphone) will facilitate mixing. Sometimes
a single cardioid microphone may be used on such programmes,
when it must be set back so as to " see " the group as a whole.
The types of acoustic for the various scenes in a studio production
will be defined by the script, and each microphone should be carefully placed to get the desired effect-preferably before the cast
assembles. Ammging for contrasting acoustics 5.om scene to scene
can often be very effective-a
sharp kntrast will frequently be
preferred to actual realism.
Some yean ago it was common practice to use several studio6 to
obtain the desired changes of scene, carrying out the mixing in a
drama control suite. This has the advantage of isolating the
scenes fiom one another, but was obviously prodigal of accommodation, and sometimes of manpower and rehearsal time. FVartime
conditions largely put an end to this technique, and ushered in the
General Purpose Studio, in which a " live " and " dead " end
are provided, and several microphones are used.
Plate 13.2 shows Studio 6~ in Broadcasting House. The microphone in the foreground was further deadened by the use of screens.
In the " live " part of the studio a C. I 2 microphone has been set
up to provide an alternative acoustic effect.
This type of studio is most successful when the " dead " end is
really dead-i.e. carpeted, curtained off, and with absorbent treatment on all wall surfaces. Provided the treatment is suffiaently
thorough, a series of usefully contrasting acoustics is possible.
All too often the need to accommodate 0th- types of programme
-music, etc.-means that the dead end is not made dead enough,
and portable screens are necessary. There screens are helpful in
two ways. Firstly, they increase the proportion of sounds reaching
the microphone after reflection over slwri paths, and mask the
microphone from long path reflections,
reducing apparent
reverberation. Secondly, by covering the surface of the screen
with rock wool, etc., some absorption of sound is achieved, with
further reduction in reverberation.
Unfortunately, portable screens suffer from the limitation that
they are relatively ineffective at low frequencies. They are in&cient as reflectors for wavelengths over about 3 ft (tiequencies
below 350 CIS),and their absorbing efficiency also falls off rapidly
in this region. Screens must therefore be used carefully if this
frequency discrimination is to be kept to a minimum.
A " tent " of screens is often constructed, to obtain an open-air
effect, i.e. no reverberation, as shown in Fig. 13.8. This requires
extra care, to avoid setting up parallel reflecting surfaces-with
consequent distortion due to standing wave resonances, which
will give a " box-like " acoustic. Superimposing records of birdsong etc., will not entirely mask the unrealistic effect of unsuitable
studio acoustics.
Bemuse of the microphone characteristics, the actual movements
of the actors in a broadcast play will often require tobe restrictedand not exactly what the context would seem to demand. Stepping
back two paces, or into the dead side, may sound like retiring into
the middle distance. When screens are being used, for example,
exits should not be made into the open studio, but always into the
non-reverberant part, so as to maintain the illusion of the particular
Occasionally a scene can be acted more or less hithfblly, in terms
of movement, with excellent results. A court-room scene played
in a large studio, with chairs laid out for judge, counsels, witnesses,
public, etc., footsteps, and so on, may come to life over the microphone much better than if performed in a dead studio with artificial
echo, and footsteps on a wooden board. The actors, too, will
usually prefer the more natural conditions.
- M S -
When using separate studios, it is a simple matter to superimpose
scenes, but the microphones in a general purpose studio must be
carefully positioned, especially if one of the scenes is a noisy one.
The acoustics of both scenes are liable to suffer during the overlap,
due to both microphones picking up the sound. If each sound
source is arranged to be in the dead angle of all microphones except
its oim, mixing will usually be simplified.
A quick return to the studio after hding out, say, a crowd scene
must be carefully rehearsed until everyone understands which light
cue means " Quiet, please! ", and which " Start next scene
Usually a series of flicks is used to obtain silence, and a steady light
for the beginning of the new scene.
A minimum distance of about three feet from the microphone
should be aimed at for solo singing. A much greater distance will
be pomible in some cases, depadiug on the relative loudness and
perspective of singer and accompaniment, and on whether the
latter is piano, orchestra, or choir.
I t is important to make the singers sound " in the same studio "
as the accompaniment, and at a natural relative distance. I t may
be necessary to discourage undue movement while singing,in order
to keep the conditions constant. The practice of" stepping back
for the high notes " will sometimes give a disconcerting change in
the apparent acoustic conditions. Certain experienced soloists
have developed ad efficient microphone technique, but generally it
is better if they keep in the same position. If isolated notes tend
to over-modulate, a slight fwning of the singds head might be
suggested rather than stepping back.
Vocalists with modern dance orchestras will usually employ a
closer technique, on account of the very loud accompaniment and
(sometimes) the very quiet crooning. A Bass Correction Unit
should generally be used, and explosive consonants watched out for
as with close speech. Using a separate studio for the vocalist, and
headphones to hear the accompaniment would simplify the separation problem, but it is not always practicable for other reasons.
Choral works are generally most sucin fitirly reverberant
conditions. The blending of voices is improved, and indeed in
some works there is little doubt that the composer has presupposed
lively acoustics when deciding the overlapping of harmonies, etc.
Moving the microphone further away-to get the best " blend "tends to make the words less distinct, and the final position is
usually at the greatest distance consistent with good diction.
With very large choirs, this limit may be reached at a distance
from which the choir is not " seen as a whole " by a single ribbon
Fig. 13.9. Large choir balona with (a) two ribbon micr0phon.s; (b) wrdioid mk~$hone
microphone, and two or three microphones may be necessary to
preserve the internal balance of the choir. The microphones must
be " phased " correctly and may be slightly angled to each other
to minimise the area of overlap. A single cardioid microphone
may be used for quite large choirs, and at a reasonable distance,
because of its wide frontal angle of pick-up. It is sometimes
necessary to discriminate in hvour of the male voices when these
are numerically in the minority (Fig. 13.9).
Close harmony groups are a regular feature in Variety broad-
casts, and present something of a problem, since they cannot be
balanced satishctorily on one side of a ribbon microphone. A
split balance can give good results, but for audience shows a cardioid
microphone is considered better from the presentation point of
view. Larger groups, and small choirs, where a " tight" (close)
balance is required, can sometimes be arranged in a V-shape on
stepped rostra, with the leading voice in fiont.
As with singers, provided the increase in bass response at distances
of less than 2 ft is avoided, microphone distance has little effect on
the actual quality from musical instruments. The working distance
chosen is therefore bound up simply with the acoustic effect required, and perspective and balance with respect to the other
instruments present. (An exception exists in the case of instruments whose mechanical action is audible-celeste, guitar, piano
accordion, and even violin-these suffer from too close a balance.)
Much more important, from the point of view of quality, is the
Mglc the instrument makg to the microphone. As previously
mentioned in connection with loudspeakers (Chapter 6 ) , a source
of sound waves is relatively non-directional at low frequencies, but
begins to exhibit directional radiation as the frequency is raised
to give wavelengths which are smaller than the source.
Exapt that radiation is more e•’Ecient from the front than the
back, the violin is reasonably omnidirectional a t low Frequencies.
At shorter wavelengths, however, the directional effect becomes
more marked, and the microphone position becomes quite critical
(Fig. 13.10).
If the higher harmonics are to be picked up, which give the
charicteristic attack and timbre, the microphone should be roughly
1 3 . ~ 0 . Polm d i c r g r y . of
violin, s k i n g inmused dmctrm(r
at high frcqumcics
a t right angles to the belly of the instrument. Note, however, that
the reproduced quality will sometimes be considered better, and
bow noise will be less, if the microphone is placed somewhat off this
A compromise is inevitable.
axis, even by as much as 45'.
Directional effects are less marked with the viola, and the microphone angle is not critical.
Excellent broadcast quality is obtainable from the 'cello, especially
in the upper register. Because of the low playing position, however,
it is important to avoid masking by other instruments. The
directional properties of the 'cello are similar to those of the violin.
The soloist in a 'cello concerto is sometimes placed on a special
rostrum. The resultant reinforcement of the sound may be an
advantage but, of course, the rostrum may possess marked resonances.
Double Bas
The double bass is om&-directional over most of its lower
register, and the principal balance difficulty is to ensure definition
in this range. Floor and wall resonances tend to increase the
blurred effect, and solid floors and walls are to be preferred. The
number of basses in an orchestra has long been known to be critical.
One more or less than the optimum number seriously affects the
balance of parts. In an orchestra large enough to employ four or
more basses, it may be advantageous to arrange them on stepped
rostra. They then present the appearance of an extended sound
source and the resultant definition may be improved. The accuracy
of the players' intonation is one of the most important factors in
getting good definition.
The harp is not particularly directional in the horizontal plane,
and the microphone angle is not critical. A balance combining
good tone and suppression of mechanical sounds from a solo harp
may be obtained with the microphone behind the player's head (see
x M 3 . Piraotorte
As its name perhaps suggests, the pianoforte covers a wide range
of dynamics. This, together with its percussive nature and richness in overtones makes this instrument a severe test of the technical
efficiency of the microphone and the programme chain in general.
The tonal qualities also depend on the studio acoustics and style of
playing, to some considerable extent.
Depending on the circumstances, a good single microphone
balance can usually be obtained within the arc shown in Fig. I 3. I 2.
The height, and angle of tilt of the miaophone will affect this, and
at normal working heights it is common practice to work towards
the tail end of the piano.
In some conditions there may be a tendency for the higher notes
to recede in relation to the lower ones. In order to correct this
error in sound perspective a second microphone can be used, placed
higher than the lid of the piano, looking down on to the top strings.
This microphone should be mixed in as required to restore the
perspective balance.
The microphone distance depends on a number of factors,
including the acoustics of the studio. For a piano recital of serious
Fig. rg.12. P d k
bo[ma firm
music, a distant technique is desirable, since a fair amount of
reverberation is called for, and distance simplifies controlling-it
being remembered that a Beethoven sonata, for example, may
exploit the fi~.Udynamic range of the instrument. Thus, with a tilt
of up to 4 5 O , the tendency is to work as far away as the given studio
will permit.
For light music, a less reverberant effect is often required, and a
distance of about 4 ft is usual. Since the height will probably
be reduced in proportion, care may be necessary to avoid the lid
The piano in modem dance orchestras is effectively part of the
rhythm section. To cut out all reverberant sound and secure
effective separation from brass, saxophones, etc., a very close balance
is usual, with the microphone a fmv inches above the treble strings
(see Plate 13.3). It is common practice to remove the lid of the
piano in order further to reduce unwanted reflections.
Upright pianos are not often used in broadcasts, but a balance
which discriminates slightly in favour of treble is usually called for
with the microphone in h n t as shown, or at an equivalent position
a t the back as in Fig. I 3. I 3.
x 3 - 4
Solo broadcasts of the clavichord and instruments of the harpsichord family may present some difficulty. The extremely quiet
sound of the clavichord dictates a close microphone position, but
care must be taken to see it is not too close, or action noise will be
troublesome. W ~ t hthe harpsichord, again, the microphone should
be distant enough to avoid picking up the mechanical action. A
programme volume corresponding to Peak Programme Meter
readings of 3 or 4 maximum for the harpsichord and 2 for the clavichord, is to be expected, and unusually high control settings should
be avoided in order to keep studio noise to a minimum.
Balance of the harpsichord with other instruments is often critical
and depends to some extent on whether a figured bass form of
accompaniment is being played or an actual melody line in polyphonic music. It is frequently difficult to obtain the correct
balance with the right perspective relationships.
The flute, oboe, clarinet and bassoon require no special mention
in this section, as their directional properties are not very marked.
The microphone should not, however, be placed in direct line with
the bell of the oboe or clarinet.
The k c h horn should be placed to face a given surface throughout in chamber or concert works, in case the player moves. In
light orchestras, reinforcement of the horns can be achieved by
backing them with hard screens.
Trumpets, trombones, etc., have considerable carrying power
in a line from the bell of the instrument. The highest overtones are
in fact confined within a narrow angle along this axis, and particularly brilliant quality is obtained from a microphone placed there
(shared by any members of the audience who happen also to be in the
" line of fire "). In solo work it is usual to place the microphone
outside this narrow angle, the quality being less " hard ", and more
consistent, if the player moves slightly while playing. Exceptions
are the trumpets in dance orchestras, and when a mute is being used.
13.8.7. Pcrcp~)dop
As mentioned in Chapter 2, percussion instruments can be
divided into two groups; instruments with definite pitch, like the
timpani (kettle drums),tubular bells, and glockenspiel; and instruments with no definite pitch, like the bass drum, cymbal and
triangle. The sounds from this latter group are extremely complex,
having component frequencies up to 20,000 cis at least. Good
results are therefore obtained only over the most efficient reproducing systems-and with very careful microphone placing.
The principal requirement here is definition, which would seem
to call for a fairly close balance; but, of course, percussion must
not be allowed to drown the other instruments. In practice, a
compromise between definition and volume is usually reached with
a single microphone, or a multi-microphone technique is used for
good definition, and the musical balance is restored by mixing.
The notes of the celeste do not carry well, and a separate microphone may be necessary, except in serious works where the accompaniment is suitably thin.
The guitar in dance bands sometimes presents a twofold balance
problem. At one stage it may be " acoustic ", when the radiation
&om the actual instrument is picked up; at another, it may be
" electric ", when the player plugs in a loudspeaker, whose much
louder tones are fed from an electric pick-up attached to the guitar.
A balance which reproduces the spacious effect of organ sound is
preferred. The microphone should be distant enough for this to
be obtained, but not so distant that the reverberation of the hall or
studio masks the clarity of the instrument. Some exploring may
be necessary to find the best position. In small churches, for
example, the layout of organ pipes is not always symmetrical, a
fact which may not reveal itself until a fidl listening test in the church
has been carried out to locate pedal notes, etc. Similarly hidden
attachments to cinema organs may exist such as tubular bells and
electrically-operated pianos and drums, which the organist should
be asked to demonstrate in case an extra microphone is required.
The small, transportable, electric organs often operate two
loudspeakers. If this complicates the microphone placing, one of
these may be disconnected, or else a layout chosen which uses one
loudspeaker for the audience, and the other for the microphone
(Fig. I 3. I 4).
Since an electronic organ generates electric currents to feed the
loudspeakers, and the microphone converts the resultant sound back
Fig. 13.14. EIrctrOnic organ, b
ing nriuophmvs p&ud lo pick up one
into electricity, it may be asked: " Why use a microphone a t all?
The alternative of simply connecting part of the organ's output to
line has been employed in certain cases, but, of course, the absence
of reverberation completely destroys the illusion of space.
Thepirmo accordion should not be placed too near the microphone,
because of key and reed noises, and usually the treb1ei.e. the
is arranged to face the microphone.
When several performers are to be grouped with respect to the
microphone, every attempt should be made to place them comfortably in their natural formation. In the rare cases where an
abnormal layout is called for, ease of performance must not be
disregarded, nor the need for the musicians to see and hear each
other easily.
In dramatic productions, distance from the microphone is
adjusted to regulate the perspective of actors. Perspective is
important in music, too-instruments should sound a t natural
relative distances-but the additional k t o r of balance applies in
this case-viz. are we listening to melody and accompaniment, or
melody and counter-melody, etc.? A complete understanding o
the relative importance of the musical parts is essential in any
studio manager who is attempting microphone balance of musical
programmes. This of course implies the abiity to read a full score
and discuss with the conductor and artistes any particular points
of balance and interpretation to make sure that these are properly
The choice of the type and number of microphones will be
influenced by the kind of music--serious or light, etc.-and
frequently by the necessity of producing particular effects in a given
hall or studio.
13.9.lr- Sangs*-
It may seem a minor point, but it must be decided at the ou&t
whether the piano is merely an accompaniment, or is of equal imporIt is safe to assume that the
tance-as in the art song (lie&).
accompanist will maintain the appropriate level, but the position
of the microphone must be right, in order that the appropriate
perspectives should be nlaintained.
In the drawing, Fig. 13.15, position (a) has the advantage that
the piano is not masked by the singer as at (b). However, this
masking effect is slight, and (b) allows much more freedom of
choice for the microphone position-both in terms of height and
distance, as shown in diagram (c).
A two-microphone balance which permits very close liaison
between singer and accompanist is illustrated at (d). Note that the
rule of non-pick-up of unwanted sources (Chapter 5) is strictly
obeyed, and the dead sides of piano and singer microphones are
directed towards singer and piano respectively.
13.9.5 Solo Instruments with Piano
Sonatas for violin and piano, etc., are usually balanced as at (b)
in the preceding section, the soloist being placed at the end of the
keyboard, and full advantage taken of the available space for
21 I
microphone placing. The novice is warned here, as elsewhere,
against close balance. A distance and height of g ft for the microphone is not unusual.
hngs at the Piano
This balance is something of a compromise. A straightforward
vocal balance is impossible since the piano would then be too loud,
Fig. 13.16. Songs at the pimro
and discriminating against the piano means that its quality suffers.
In practice, the piano lid is put on a short stick, and the microphone
tilted, as shown in Fig. 13.16, to reduce piano volume. A separate
microphone may be mixed in to achieve better piano balance.
Two common arrangements are illustrated in Fig. I 3. I 7, and it
will be seen that a single microphone is used. The position B is
usually prefemed, as greater liaison between the pianists is possible.
If the use of a single microphone gives too reverberant a soundsay, in light music-two may be tried as at C and D. With this
layout, the lid of one piano is sometimes removed completely, but
then seuns to be little justification for this.
A moderately live acoustic is nowadays aimed at for broadcasts
of chamber music. I t is imperative that no shift in relative perspectiveshould take place during playing, and so a singlemicrophone
is used, at a fair distance, and the players are grouped for ease of
playing and equal balance (Fig. I 3. I 8).
If a more intimate effect is required it may be necessary to use
more than one microphone in order to preserve the relative perspective and balance between instruments.
If a piano or harpsichord is included, a second microphone (at a
fixed level) may occasionally be required, as shown in the diagram.
Variation of the " mix " of microphones is hardly ever necessary
in chamber music.
Light Music Ensembles
A single microphone may well suffice for light music groups, plus
a piano microphone if required.
The leader (first violin) of such a c o m b i t i o n may remain
standing, as shown in Plate 13.4, and may require an extra microphone for quiet solos. Modern arrangements, however, are tending to need a multi-microphone technique and two such typical
layouts are shown in Fig. 13.19.
For larger light orchestras, modem scoring often makes a multimicrophone balance obligatory. The wide variation in the relative
loudnesses demanded from the instruments in modem orchestrations may be reproduced only by the mixing of perhaps four or more
Plate 13.I . Discussion wilh fiue speakers, showtng me of c . 1 2 microphone slung over table
Plate 13.2. General view ofStudio 6.4,Broadcasting House, ~howing'~live"
and"dead9' ends
Plate 13.3. Close piano balnnre
Piale 13.J .
Smal/ fighl music combination
Plate 13.5. Multi-microphone light m u m balance
Plate 13.6. Use of u single I;. 12 n~icrof)hone,/or!he Royal I~/~iIhannonic
Orrhe.rtra in hlairln Vnlr. .Sttdio
Plate 13.7. The Dit Disley Quintet
Typical microphone placing5
for dance band inttrr~merrls
microphones. The layout of the BBC Concert Orchestra in the
Camden Theatre is shown in Fig. 13.20.
Plate 13.5 shows a complicated microphone balance in which
the number of microphones exceeds the number of musicians.
The Classical Orchestra
The orchestra used up to about the year 1800 (Haydn, Mozart,
etc.) is made up roughly as follows:
8 first violins, 6 second violins, 4 violas, 2 'cellos, 2 basses, 2 flutes,
2 oboes, 2 clarinets, 2 bassoons, 2 horns, 2 trumpets and timpani.
The layout of the orchestra-strings in front, then woodwind, and
brass and percussion behind-is logical, since the carrying power of
the instruments increases in that order. Also, in terms of the ratio
of direct sound to reverberant sound, this layout gives best results.
The attack and clarity of the string tone is best reproduced if direct
sound is strong, whereas the blending of the more sustained brass
tone is improved by reverberant conditions.
At least three possible seating arrangements of the strings are
available to the conductor, reading from left to right :(a) ~ sviolins,
violas, 'cellos, 2nd violins
(b) 1st violins, 2nd violins, violas, 'cellos
(c) 1st violins, 2nd violins, 'cellos, violas
From the microphone point of view (b) and (c) are perhaps to be
preferred, since the violin skctions are together.*
A single microphone should be used, placed to reproduce faithfully the internal balance of the orchestra under moderately reverberant conditions (Fig. I 3.2 1 ) In cases where slight discrimination in favour of the upper strings is desirable-perhaps because
they are few in number-a microphone at A or C will be chosen,
and not B. Assuming a normal angle of tilt, A will help by its
nearness to the upper strings, and C by its orientation. C will give
better discrimination against trumpets and percussion in the layout
shown, should there be any tendency for them to sound too loud.
The microphone distance is of primary importance. It must be
far enough away to " see " the orchestra as a whole, and reproduce
the internal balance, but not so distant that reflected sounds blurr
the orchestral tone or articulation. In practice, depending on the
For stereophony (a) may be preferred, in order to achieve the antiphonal
effects desired by some composers.
4 4 0 3 8 ON
* 4 0 3\8 ON LAZY' ARMA G
~ I ~ A ~ ~ A ~ ~
Fig. 13.19.(a) Microphone layout of "Gland Hotel"
hall or studio, distances from about I 2 ft to 30 ft are used with ribbon
microphones and slightly less with cardioid.
13.9.9. The Modern Symphony Orchestra
From Beethoven (I 770-1827) onwards the orchestra has grown in
size and variety of tonal colour. The modern composer makes use
of an orchestra roughly as follows :20 first violins, 16 second violins, 12 violas, 10 'cellos, 8 basses,
3 of each of the following-flutes, oboes, clarinets and bassoons,
with the third player doubling on piccolo, cor anglais, bass clarinet,
and contra-bassoon respectively, 4 horns, 3 trumpets, 3 trombones,
I tuba, 2 harps, celeste and 3 percussion players with a variety of
Fig. 13.19.( b ) Microphone layout of "Chad9 Katr Novelty Scxter"
All the remarks in the previous section apply here-with the
added difficulty of the large area taken up by the players. The
brass section, for example, may be as much as 30 ft from the front of
the orchestra.
If a microphone were placed closer than 30 ft from the strings,
the perspective (i.e. ratio of direct and indirect sound) would vary
seriously throughout the orchestra. In circumstances where a
distant microphone position is possible, the " out of focus " effect
is overcome, as in Plate 13.6.
It must not be supposed that all sections of an orchestra should
sound in the same perspective. In the studio, live, the brass and
\voodwind in fact sound more distant than the strings. This
Fig. 13.20.
Microphone layout of BBC Comt Orchestra in the Camdm T/uabe, LoLondon
d e e d
6 d 8 6
Fig. 13.21. A typial arrangmni of the classical orchestra, showing three akma&
microphone positwtu-A, B and C
natural effect of depth must be preserved over the microphone, in
order to give " depth " to the orchestra.
The Concerto
Again, it is the true balance which is to be reproduced, and a
single microphone should be used when possible.
A solo piano will usually be placed in front of the conductor, or
slightly to his left, and other soloists-violin, horn, clarinet, etc.at position X, in Fig. 13.22(a).
In cases where the concerto microphone is in a different position
from that used for orchestra alone, it is very important to ensure
that the orchestral perspective and the apparent acoustics are substantially the same on either microphone. The listener should hear
lh c m m t o - ( a ) nmrrml arrangement, wing Jingk
microphmu, and (b) kmmicrophone b
the same orchestra playing in the same place in orchestral and
concerto items. When some form of compromise has been necessary, due perhaps to lack of space, the changeaver of microphones
should take place as imperceptibly as possible, or it may be preferable to take the concerto on the orchestral microphone, aided by
very little of the output of a close microphone, as for singers in
Section 13.8. I I.
In Fig. 13.22(b), the piano has been placed to the side, and two
microphones (B and C) are needed for the concerto. B is closer than
A, which is used for orchestral items, since the added orchestral
reverberation picked up by the piano microphone C must be compensated for. If the concerto, following an overture, symphony,
etc., has an extended orchestral introduction (as in many classical
2 18
concertos), it may be preferable to remain on Microphone A for
this, and change over to B plus C just before the entry of the soloist.
The change in musical colour at this point will help to disguise the
change in orchestral balance.
Needless to say, the above device may be unnecessary if applause
and a long announcement separate the items, and will be impossible
if the orchestral introduction is very short. The settings of faders
B and C should, of course, be pre-arranged at rehearsal, and maintained during the whole work, as no shift in apparent perspective
must take place.
songs with OrehesfrP
In a concert hall, singers will normally take up a position at the
front of the platform, and a subsidiary lower microphone will often
Fig. 13.23. Smgs with cr&Ira.
(a) shows one method of tilting the so&tls
with the singer's microphone
mirrophone, and (b) shows tht normal studw b&
at right a n g h Lo the orchestra
be necessary. Careful tilting will help to minimise perspective
distortion on orchestral pick-up, as in Fig. I 3.23 (a).
In studio performances, singers may be placed on the conductor's
left or right, as preferred, and the orchestral and vocal microphones
may then be angled for easy mixing, in accordance with the
principles set out at the beginning of this chapter (Fig. 13.23(b) )
choirs with orchestra
It is rare that choir and orchestra can be picked up satisfactorily
on one microphone. The limiting factors are intelligibility fiom
the choir, and definition from the orchestra.
In studio performances, when the numbers are not too great,
the choir can be arranged at right angles to the orchestra-on the
Fig. 13.24. Choirs with orchtra. (a) shows Mmral sfudio b a k , an$ (b) shows
split balance on large choir behind orchestra
conductor's left or on his right-with a separate microphone (see
Fig. rg.rq(a).)
When the choir is placed behind the orchestra, care is necessary
to avoid pick-up of parts of the orchestra. The angle of tilt of the
choir microphone is quite critical in this case. Very large choirs
of 200 voices or more may make several choir microphones necessary,
and diction is likely to be indistinct.
One or more ribbon microphones angled as shown in Fig. 13.24(b)
have often been used successfully with large choirs. The use of
cardioid microphones for choirs has already been mentioned above.
The layouts for brass band broadcasts will usually conform to the
pattern shown-namely, first and second cornets playing towards
e e
a a
6 8
9 d
9 6 6.
Fig. 13.25- Laput of stmdard bras band
Fig. 13.26. Militmy band
Two cxamplcs o
band balance with one minophone
22 I
each other, backed by sopranos and trombones respectively, with
horns, euphoniums and basses making up the third side of a square
(see Fig. 13.25).
This arrangement permits good definition on horns and euphoniums and prevents their being swamped by the direct tone of
cornets and trombones. The basses, of which there are usually
2 Eb and 2 Bb instruments, are very important from the point of
view of balance of the musical parts.
Comet solos present no balance problem, but solos on the trombone or euphonium are a little difficult, because of the inherent
fig. 13.28. Ekmnplc of microphone arrangement fw
Edmwrdo Ros's band
weight of the accompaniment. In some cases a "spot " microphone may be used, or the soloists may be asked to come to the
front of the band.
Artificial reverberation is sometimes used for brass bands when
a close balance has been used for good definition, and natural
reverberation is felt to be lacking.
A roughly three-sided layout is used again, with flutes, oboes,
clarinets, and saxophones taking up the front rows. The percussion is more elaborate than the bass-drum, sidedrum and triangle
of the brass band; timpani and xylophone are quite common.
Special attention should be given to the percussion effects at
rehearsal, as precision of balance is required (Fig. 13.26).
Until a few years ago, a dance band played musical arrangements
which were straightforward. The melody line was firm, and the
accompaniment held at a level which presented a balanced sound
to the listener on the spot. It followed that a one-microphone
balance was possible, with occasional use of " spot " microphones.
Fig. 13.29 Microphone arrangement
for Ted Heath's band
Arrangements of this kind are occasionally met, and a layout based
on Fig. 13.27(a) may be used. Another example of one microphone being used is the Victor Sylvester dance band, shown at (b).
When greater separation between sections of the band is required,
and with the unnatural balances of up-to-date arrangements, e.g.
celeste melody against a brass accompaniment etc., a multi-microphone balance is obligatory. A separate microphone is used for
Fig. 13.30. Minophont arrmqmunts for a traditional band and a modem ja.& band
Fig. 13.31. dficrophont arfangmrmt for
a jazz
C/l z
9 9
Fig. 13.32. Anangcmcn~fm a variety orchestra
each section of instrument., and careful angling, together with close
positioning, serves to reduce out-of-phase effects to a minimum.
Screens may be used to increase the separation between the various
sections of instruments, and under some circumstances a " tent " of
screens may be used for a vocalist.
Some typical arrangements of microphones used in jazz group
and dance band balances are shown in Figs. 13.28 to 13.31 and
Plate 13.7.
Variety Orchestras
The variety orchestra is a special case in that the music played
may vary from the popular classical repertoire to typical dance
band numbers, and so a balance technique capable of adjustment
to these two conditions is essential. A typical layout is shown in
Fig. 13.32 and Plate 13.8.
The volume limits between which satisfactory broadcasting or
recording are possible in a given system define what is called the
dynmnie rrmgc of the system. In disc recording, for example, the
groove spacing (dictated by economy of playing time) sets an upper
limit to the recording amplitude; at the same time the lower limit
will be &zed by considerations of surface noise. Between these
limits we are normally left with a dynamic range of not much more
than ?o dB.
In ';he programme chain for broadcasting, a number of other
facton enter into the problem, such as transmitter modulation
limits, level requirements in line circuits, etc., causing a hrther
restriction of the dynamic range.
In the domestic services of the BBC, an average range of little
less than 30 dB is aimed at (programme meter readings &om I to 6,
for example, correspond to a range of 22 dB).*
In the European and Overseas services, a narrower range is used,
to allow for the extra noise and interference experienced on shortwave reception.
Comp'ession is the name given to the process whereby the fluctuations in programme volume from a studio are restricted (automatically or by manual control) to conform with the narrower dynamic
range of a given system. The diagram, Fig. 14.I, shows the approximate range of sounds generated in speech and in music. It will
be seen that broadcast talks, which occupy a range of about 20 dB
In primary service areas listeners, particularly on V.H.F., may well be able
to use a dynamic range in a- of this. After all, there is an infinite number of
decibcls between o and r on the P.P.*M. The restriction is only necesary when
local noise is high.
may escape " compression, whereas music (a range of approximately 60 dB) may require a measure of control to keep within the
dynamic range of the system.
The drawing also shows the wider frerange of music, compared with speech, which explains why attenuation distortion is more
serious (because it is likely to be more noticeable) on music than on
speech (see Chapter 12).
Reason for Manual Control
It is possible to introduce compression automatically, using a
limiter circuit, but results are usually disappointing h m the
artistic point of view. The musical expression, or light and shade,
is a function of the fluctuations in programme volume, and in
particular of the sharp contrasts in volume. These tend to be
flattened out in automatic compression.
I n manual control of the programme volume the studio manager
or engineer is often able to anticipate contrasts in the level, and effect
the necessary compression without seriously distorting the musical
expression. He is assisted in thus imperceptibly controlling the
volume by two characteristics of the human ear:142.1.
(a) The ear cannot detect small changes in level (steps of 2 dB or
(b) The ear cannot compare accurately levels which are separated
by a time lapse.
Although there are nearly as many problems in controlling programme volume as there are types of progmmrne, a rough procedure
is outlined below as a general guide.
(a) Choose a studio layout and microphone position which will
avoid, as far as possible, unnecessarily great contrast in volume
(e.g. placing of soloists, effects microphone, etc.)-the aim being
to reduce controlling to a minimum.
(b) Estimate early in the rehearsal the average setting of the main
control potentiometer; all adjustments will then be made away
from this setting, which is regarded as " home
(c) Note at rehearsal any points in the programme which exceed
the permissible range with the control potentiometer at the
" home " setting; and estimate the " away " setting that will be
(d) Anticipate these points by gradually fading up or down in
advance (single stud adjustments at intervals of a few seconds will
usually be imperceptible), so that the sharp contrasts are maintained, and not levelled out.
As an example, Fig. 14.2 shows a sudden peak followed by a
gradual diminuendo. The control shown involves three down-
Fig. 14.2. Ad-
control porcntimnctm in manual control of&
w a r d steps of 2 dB before the peak, and five upward steps during the
quiet passage.
Although the above remarks apply most obviously to music programmes, they have a teal application in dramatic productions also.
Pistol shots, to take one of the many examples that spring to mind,
may be disappointing, not because of any real lack of pistol quality.
I t may simply be that they don't sound loud enough. If the preceding
dialogue peaks 5, for example, and the explosion peaks 6, it will seem
scarcely louder than the speech. Better practice would be to
reduce the dialogue volume gradually in advance--say, todna&
then establish a contrast of about 10dB. In the same way, contrast
should be engineered between noisy crowd scenes and quite dialogues, etc., otherwise the crowd will not sound loud engouh.
The Limiter
However carefully manual control is carried out, there is always
the risk that sudden peaks may exceed the permitted maximum. A
limiter is therefore inserted into the programme chain before the
transmitter or disc recording machine, as a means of protection
against damage and distortion. The limiter takes *e form of an
amplifier whose gain is fixed for all normal levels of input. When
the input exceeds the permitted maximum, the gain of the limiter
is instantaneously reduced, so that a characteristic such as the
dotted line " x " (see Fig. 14.3) is followed, and the output level
cannot exceed 6. On isolated peaks, the recovery time of the
limiter is very short, and the action is virtually fiee of the usual
types of distortion.
The limiter, however, by its very action, produces abrupt distortion of the original dynamic range, and this can be very objectionable if the limiter is allowed to function on a number of successive
peaks, or, indeed, over any long period of time.
We now come to the question, " Should music peak more than
speech and, if so, by how much?"
The drawing in Fig. 14.1 indicates that music extends to higher
intensities than speech-and we usually expect loud music to be
louder than loud speech. As a general principle, therefore, it
would seem natural to arrange that music peaks higher than speech.
However, this is not always a desirable state of affairs; so much
depends on the type of programme and the type of audience for
which it is intended.
There are two basically different types of audience for a radio
programme, firstly, the serious listener, that is the listener who does
nothing else but listen to the programme, frequently on high quality
equipment, and secondly, the listener who prefers his radio as a
background whilst he is performing some other activity. Some
programmes may be considered as solely for one or the other, and
suitable treatment can be arranged to cater for these. Other programmes may be listened to by both types of audience and in this
case suitable treatment is more difficult.
In the k t of the cases, the natural relationship referred to above,
where loud music should sound louder than any accompanying
speech, is usually preferred. Such a listener, for example, when
Fig. 14.3.
listening to an orchestral programme, will probably adjust hisvolume
control to give a loudness approaching that which he would have
heard in his favourite seat in the concert hall, and in this case he
will require announcements within the concert, and, preferably the
opening announcement to the programme following the concert,
to be kept to a suitable low volume.
In the case of the background listener, he will adjust hisloudspeaker to a fairly low level, such that the music is clearly audible
but does not interfere with normal domestic conversation. Thus
it will be by no means as loud as the previous case. If the announcements are then broadcast at a lower volume than the music, they
may well fall below the conversational level and so will be inaudible.
This will cause the listener to adjust his volume control frequently
in order to keep the levels right for him, resulting in frustration and
It seems quite clear that although the two programmes quoted
above consist of speech and music, a different studio technique is
necessary in so far as the adjustment of the relative levels of speech
and music are concerned. I t is the function of the pmddction
department to determine in which category a particular programme
is placed and to instruct the studio staff accordingly.
It has been found convenient to apply the " background listening"
conditions to many programmes taking place in the early morning,
and in same cases up to tea-time, and also to programmes in the
early evening, when people are returning h m work listening to
their car radios, and housewives are preparing an evening meal.
Programmes in the " serious listening " group w i l l occur in the
and, more particularly, during the evening tmmnkion
. .
An account of investigations into listeners' p d m m in this
matter appeared in the BBC Quarterly for Spring 1950. The tests
centred on the reactions of 60 people to a number of junctions
between music and speech programmes at several relative levels.
Programme junctions were used becausc most of the complaints
from listeners about volume relate to the connast between consecutive programme items. It is possible to summarise the results of
the test in the following recommendations:-
Speech following music to be 4 dB down.
Music following speech to be 2 dB up.
3. Speech following speech to be within f 2 dB.
4. '' Bow Bells " interval signal to be rg dB below speech.
From these recommendations the following points arise:(a) The discrepancy between I and 2 is not explained, but may be
allowed for by starting a music programme 2 dB above speech,
and then making up the other 2 dB when artistically possible in
the music.
(b) Item 3 confirms the fact that intensity changes of 2 dB or less are
(c) Incidental music in drama and variety programmes is not, of
course, covered by the above summary. In fact, there is considerable evidence that listeners prefer music within such programmes
to peak several dB below speech. In dramatic productions, the
music should never be higher, but certainly lower than if it were a
concert item. Similarly, if the musical items are controlled to
the same peak volume as speech in variety productions, they will
often appear to be too loud.
It will have been noted in the above discussion that the l o h s
of the sound was referred to, rather than the P.P.M. reading. It is
essential that for the purpose of varying such loudness, the studio
manager relies upon his ears when listening to a loudspeaker adjusted
to a suitable level. It is possible that two sounds which to the ear
sound vastly different in loudness, may give similar readings on the
P.P.M., and so clearly this instrument cannot be used to give an
indication of loudness. It does, of course, give an accurate indication of the peak volume permissible in order to avoid amplifier and
transmitter overload and consequent distortion. Likewise, it gives
a good indication of the lower transmission limits below which the
volume should not go if the programme is not to be in danger of
being swamped by noise.
The studio manager therefore, has a dual responsibility in this
matter. Firstly, he must send to the listener the sequence of loudness which best represents the artistic content of the programme,
bearing in mind the particular audience for which the programme is
intended. Secondly, he must control the programme volume
within the technical limits of the system.
INmany progranmws, especially drama, operatic and light entertainment programmes, various sound effects are often required.
These sound effects can be produced by three basic means:-" spot "
effects, recorded effects and radiophonic effects.
In the early days of broadcasting it was not practicable, because
of the less advanced equipment, to produce convincing effect sounds
on gramophone records. More often than not such a recorded
effect would sound unconvincing, if only because the disc records
of that time had rather high surface noise. Furthermore, and this
problem remains at the present time, it was not possible to synchronise recorded sounds easily with live action in the studio. Because of
this, a number of effects are actually created in the studio, and these
are known as " spot " effects (Plates I 5. I -3). They range from the
more obvious sounds such as footsteps, doors opening and closing
etc., all of which are the actual sounds of the respective pieces of
equipment, to elaborate artificial effects, where the most unlikely
objects may be pressed into service to create the required sounds.
These last are seldom used in dramatic products at the present time,
but still exist in some light entertainment programmes for comic
For those interested in experimenting with unusual spot effects,
a number are listed below:(a) Seawash: a handful of lead shot placed on the skin of a singlesided drum and gently rolled from side to side.
(b) Avalanche: a quantity of potatoes placed in a large bass drum
and rolled from side to side.
(c) Squeak of car brakes: an inverted tumbler slid on a sheet of
(d) Walking in snow: twisting a roll of new cotton-wool very close
to the microphone.
(e) Machine gun: lead shot in a single-sided drum, vellum tapped
with drum stick.
(f) Railway train: roller skate moved over stiff wire attached to
(g) Flight of arrows: swish of cane close to microphone.
(h) Creak of door: string attached to door handle, cloth impregnated with resin drawn along string as door is opened.
(i) Squeak of old inn sign: prongs of fork scraped to and fro round
china plate.
(j) Windscreen of car splintering: wafer biscuit crushed close to
By far the majority of natural sound effects used in modern productions are specially recorded on tape, or more probably on disc.
It has been found that 78 r.D.m.
discs are most suitable for this work.
since accurate location of any point on the recording is possible
with suitable equipment. Fine-groove discs a t 334 r.p.m. can be
used for sustained backmound
sounds. such as crowd and restaurant
noises, etc. Tape effects are used in many European countries and
to some extent in the BBC, but they require a rather different
techniaue in ~roduction. When discs are used for effects it is
possible at any time during rehearsal immediately to change an
effect for a more suitable one in the light of production experience.
With tape, whilst this is obviously possible, more time is needed
since the old effect will have to be removed from the composite
effects tape and a new one inserted.
RCCOtdiPg of S a d Meets
Recording of sound effects is most usually achieved by means of
batterysperated tape recorders, the tapes so obtained being later
processed on to disc. The most important point to bear in mind
when recording for sound effect purposes is that any extraneous
noises present at the time of recording will, in all probability,
sound quite incongruous in the situation for which the sound is
intended, and indeed, even supposing such sounds were not out of
place in a particular production the recording would almost certainly be kept for a later date, when such noise effects would be
quite unsuitable. In most cases when recording a sound effect, it is
necessary to get as close to the sound as is conveniently possible, in.
order that background noises can be kept to a minimum. If the
effect is required to sound at a distance in a given production, this
can usually be achieved in the studio.
15.2.2. Repraducdon of Recorded Effects
When recorded effects are to be reproduced into a programme,
three methods are possible. Firstly, the output of the reproducing
machine can be directly connected to the channel fader on the studio
desk and the effect mixed in as required. This is the normal method
of operation but can be a disadvantage under some conditions
when it is necessary for artists to time their speeches to coincide with
parts of the effect. In this case, two alternatives are possible. The
recording can be reproduced, not directly into the programme chain,
but on a loudspeaker in the studio, a system termed Acoustic Effects
Reproduction (A.E.R.). By this means, the effect sound is audible
to the actor in the studio and can be picked up by his microphone or
another suitably placed. Thus relative perspective and timing
between speech and effect can be controlled. The third method is
a combination of the previous two, whereby the material is fed
directly into the programme chain and also to the A.E.R. loudspeaker; the proportion can be varied to suit dramatic requirements.
With the development of the more imaginative type of radio
production, it became necessary to invent sound effects which
would suggest, in the imagination, particular emotional ideas
rather than actual everyday sounds and to this end a new effects
technique has been developed which has become known as Radiophonics. This term is not to be confused with the French " Radiophonique " which is merely an adjective applied to anything
concerned with radio.
In the present sense, the result is a combination of the French
" Musique Concr&te", the German and Italian electronic music,
and straightforward sound effects. The sounds are created in the
Radiophonic Workshop of the BBC by manipulation of tape
recordings using a number of different methods (Plate I 5.4).
15.3.1. Simple Manipulations
The simplest form of tape manipulation is probably that of
changing the speed of the tape. If a natural sound, such as one
stroke of a bell, is recorded, and then the tape speed is reduced, a
complete change in the character of the sound results. This is not
only due to the reduction in pitch, but because the harmonic structure of the sound is altered by the speed reduction process. It is
also possible, of course, to play such sounds in reverse, thereby
converting the attack of the bell into an abrupt stop at the end of the
sound. A sequence of such bell notes can be hrther altered in
character, by suppressing the attack at the beginning of each note,
either by means of a volume control or by physically removing it by
Fig. 15.1.
Ta& feedback
cutting it out of the tape. Obviously the number of sounds that
can be treated in this way are limitless as also are the various
adaptations of this simple treatment.
Any of the methods of adding artificial echo or reverberation
wred in the studio can be added to the sounds made by tape manipulation. Further effects are possible by yet another use of the tape
machine. If the replay amplifier output, on a given tape recording
of the sound which is to be treated, is fed back via an attenuator to
the record amplitier, the physical separation between the record
and replay heads of the machine will produce a repetitive effect
akin to an echo, the amplitude of which can be controlled up to
oscillation point by means of the attenuator (Fig. 15.1). The delay
can be further increased and extra echoes added by allowing the
tape to pass from the first machine and across the replay heads of
any number of other machines before being taken up on .the last
Fig. r5.2.
machine. The outputs of all replay heads in the chain can be
fed back in varying amounts to the first record head (Fig. 15.2).
Having produced a number of basic sounds, by the means outlined above, these can be cut and joined into loops of tape, which
when placed on a machine will play continuously for as long as
required. A number of such loops on different machines can then
be combined on a mixing desk to produce a " multi-layered "
montage of sound.
So far all the equipment described has been of the basic variety,
namely tape machines, tape and editing material. In addition to
these, oscillators are required for the production of accurately
determined sine wave tones (Plate 15.5). These tones can be
combined in differing frequency and proportion, and constitute
another source of basic sounds; they can then be added to the
sounds produced from natural sound sources and manipulated in
the same way. The pure sine wave need not be used; it can be
passed through a square-wave shaping device which will produce
a sound rich in harmonics. This can then be modified by means of
filters to produce further different sound colours (Plate 15.6).
Plate 15.I . Studw Managers making sound eJecfs
,/or " .Clorte d'Arfhur "
Plate 15.2. Studio Manager " cutting lht grass "
for a garden scene
Plale 15.3. The
" spot " eJects
Plate 15.5. Bonk of oscillators, showing use of freqmiqy standord (on right' to set up
occurate rombinotwns of tones
Plate 15.6. Filter, c~oriablefrom n-/;o d B otfenunlion in
A octar,e sle,bs
Filters can be of several types; high pass, low pass or band pass,
and can have varying degrees of attenuation.
Yet another source of basic sound used in radiophonics is the
white noise generator, a device which produces equal intensity of
sound over the whole of the audible frequency spectrum, a sound
which if heard on a loudspeaker is akin to that of escaping steam.
Since this sound contains all possible frequencies in equal amounts,
sharp filtering can produce bands of noise which can have characteristic pitch, and these can be used in a similar manner as before.
The above description gives an outline of some of the processes of
producing sound effects by radiophonic means. Obviously, the
potentialities are limitless and new sources are continually being
devised. The BBC's Radiophonic Workshop produces such effects
for television and sound Dromammes and the demand in both
directions is increasing.
isuhoped that the brief description in
this chapter will make it possible for others to attempt the creation
of effects in this wav.
One might have' thought that such effects are new, but the
following extract from " The New Atlantis " by Francis Bacon,
written in 1624,seems to suggest it may a11 have been done before!
" Wee have also Sound-houses, wher wee practise and demonstrate all Sounds, and their Generation. Wee have Harmonies
which you have not, of Quarter-Sounds, and lesser Slides of
Sounds. Diverse Instruments of Musick likewise to you unknowne,
some sweeter than any you have; Together with Bells and Rings
that are dainty and sweet. Wee represent Small Sounds as
Great and Deepe; Likewise Great Sounds, Extenuate and
Sharpe; Wee make diverse Tremblings and Warbling5 of Sounds,
which in their Originall are Entire. Wee represent and imitate
all Articulate Sounds and Letters, and the Voices and Notes of
Beasts and Birds. Wee have certaine Helps, which sett to the
Eare doe further the Hearing greatly. Wee have also diverse
Strange and Artificiall Ecchos, Reflecting the Voice many times,
and as it were Tossing it: And some that give back the Voice
Lowder than it came, some Shriller, and some Deeper; Yea some
rendering the Voice, Differing in the Letters or Articulate Sound,
from that they receyve. Wee have also meanes to convey
Sounds in Trunks and Pipes, in strange Lines, and Distances
IN recent years considerable interest has been aroused in the
possibility of stereophonic reproduction under domestic conditions.
So fir, only a limited amount of broadcasting on an experimental
basis has been possible owing to various transmission difficulties,
but numerous recordings both on disc and tape are available for
purchase on the commercial market. I t is felt that this book would
not be complete without some reference to the various systems of
achieving a stereophonic sound picture and some description of
equipment and studio technique designed for this. For the
purposes of this chapter we shall consider only domestic stereophony,
that is, two-channel stereophony, since it seems unlikely that any
system using more channels than this will be economically or
technically possible for some time. Multi-channel systems are used
in cinema stereophony but these will not be discussed here.
I t is probably fair to say that ever since the invention of the
microphone and the headphones, and later still the loudspeaker,
attempts have been made to achieve reproduction which would
simulate the spatial distribution of sound normally heard in everyday life. The earliest recorded experiment was towards the end
of the last century when two telephone-type microphones were
placed in the footlights of the Paris Opera and their outputs
separately connected to pairs of headphones. The report of the
experiment said that listeners were able to locate the position of the
performers on the stage. Many other experiments were conducted
before an acceptable system of stereophony on loudspeakers was
achieved, but two series of experiments, both carried out in the
Ig30s, can fairly be said to have pointed the way to present techniques. One of these was conducted by the Bell Telephone Laboratories in the U.S.A. and resulted in a number of demonstrations,
notably the transmission over land lines to Washington in rg34.of a
concert performed in Philadelphia. Three years earlier, in
England, A. D. Blumlein, of the Columbia Graphophone Co. Ltd.,
also conducted experiments culminating in the filing of British
Patent No. 394325, which describes a complete system of stereophonic recording and reproduction from microphone to disc: this,
although not practicable at that date, has become the basis of the
modern stereophonic long-playing record.
Before discussing these two main systems in detail, we will consider
how the human brain interprets sounds arriving at the ears in order
to determine the position of those sounds. Early theories on the
subject suggested that it was the difference in loudness at the two
ears which gave the required information, the head having a
shadowing effect on sounds arriving from one side. Lord Rayleigh,
in 1896, in his " Theory of Sound ",pointed out that this intensity
difference would only be significant at the higher frequencies,
above 700-~,oooc/s, where the wavelength of the sound is shorter
than the distance between the ears. He suggested that the low
frequency directional information might come from the difference
between the time of arrival of the sound at one ear and at the other.
More recent experiments on the subject, notably those by Cherry
and Leakey at Imperial College, London, confirmed the low
frequency suggestion but indicated that the effect at the high frequencies was also produced by time difference and not by the
amplitude difference. In this case, however, it was not the
difference in time of arrival of the high frequency waves themselves
which gave the required information. In natural sounds, almost
all high frequency components are varying at some lower frquency
(e.g. the " warble " of bird song, the vibration a violinist imparts to
his notes, and the beats between various sounds) and it was this
" modulation envelope " which supplied the information to the
brain. " Pure " high frequency sounds, tones, for example, were
almost impossible to locate.
The experiments further showed that the effect of inter-channel
time difference is to blur the sharpness of the reproduced image.
It would seem likely therefore, that a loudspeaker system of
reproduction which could reproduce at the ears of the distant listener
the interaural time difference that would have been heard by a
listener in front of the original sound source, would achieve the
desired result. Both the Bell and Blumlein systems of stereophony
attempt to achieve this.
The Bell system of stereophony has been termed the " wavefront "
system because its original idea was a reduction of an ideal situation
where a " curtain of microphones ", infinite in number, was
Fig. r6.r.
Thrcc spaced minophotza-three
supposed to be hung in front of the sound source, each microphone
being connected to a corresponding loudspeaker in a similar
" curtain " in the listening room.
In this imaginary case the whole
wavefront of sound leaving the sound source, should be continued
from the loudspeakers in the listening room. Obviously such a
hypothetical " sound-curtain " would be quite impossible and so the
Bell engineers reduced their number of microphones and loudspeakers to three and later to two. For the purposes of the experiment, a " caller " moved about to a number of predetermined
positions in front of three omni-directional microphones arranged
in line in a studio. In a small auditorium three loudspeakers, one
connected to each microphone in corresponding positions on the
platform, reproduced his voice, and observers were required to mark
the apparent position of the " caller " on a plan (Fig. 16.1(a) ). A
sound-transparent curtain was hung in front of the loudspeakers to
assist the illusion. To complete the experiment the " caller " made
his moves behind this curtain in the listening room, no electronic
equipment being used. I t was found that fair accuracy of positioning
across the stage was possible, but that the positions in between any
two microphones produced an effect in the listening room of the
" caller " moving upstage away from the observers, although he
did not in fact do so. This might have been expected, since he
was then farther from either microphone. If the " caller " walked
from one side of the stage to the other in a straight line, it appeared
from the listeners' point of view that he walked in a curve slightly
upstage at first, coming down to the front again by the centre
microphone and then up again and down towards the microphone
at the other side (see Fig. 16.I (b) ). If the centre microphone was
removed, thus giving a two-channel system as domestically desirable,
this effect was rather worse, there being a distinct recession of sounds
in the centre of the stage. This simple system, using two omnidirectional microphones, was the basis of many early stereophonic
recordings and gave rise to the criticism that such stereo had a
" hole in the middle
An attempt to fill up this hole has been
made by reintroducing the centre microphone and connecting its
output not to a third loudspeaker, but equally to the other two,
thereby producing equal outputs from the two loudspeakers, a
condition that would give a centre image. This is more or less
successful, but tends to produce two lesser " holes ",one-third and
two-thirds the width of the stage, unless care is taken in microphone
placing (Fig. 16.2).
Many American recordings made at the present time use this
system as a basis although other reinforcing or " spot " microphones
may be added to assist in the general effect. This system is capable
of good spatial distribution of sound but may not give " pin-point "
positioning of component parts of a complex sound sowce; for
example, an oboe player in a symphony orchestra. I t will, however, give good spatial representation to the various sections of an
orchestra: for example, it will separate wood-wind from strings or
16.s.a. Caincidcpt Microphone System
To turn now to the Blumlein system. In addition to describing
a spaced microphone system using omni-directional miuophones,
similar in many ways to the Bell system, Blumlein also described a
system using two very closely placed directional microphones. This
system has become known as the " coincident " microphone system,
since ideally the microphones should in fact be in the same position.
This, of course, is not possible and in practice they are mounted as
close to one another as possible. The operation is as follows: if
we refer back to the theories of directional hearing discussed above,
we will see that in order to produce a similar effect in space to the
original sound, we have to create in the ears of the listener the same
time differences that he would have heard in a corresponding position in front of the original sound. It can be shown that in order
to do this using two loudspeakers all that is necessary is to feed the
loudspeakers with sounds in differing amplitudes, this amplitude
difference being proportional to the angle at which the sound arrives
at the studio microphone. Two figureof-eight microphones
mounted with their main axes at go0 will produce differing outputs
depending on the angle of incidence of a given sound to each microphone. Such a microphone pair, therefore, will provide the
necessary transformation between position in space, and amplitude
difference between loudspeakers, to achieve the desired effect
(Fig. 16.3). Microphones of other directional characteristics can
also be used, depending on the type of pick-up required. In this
case good positional accuracy is possible, limited by the acoustic
conditions in the studio and more particularly in the listening room.
Choiceof Systams
Which of the above two methods is to be used in a complete
system of domestic stereophony is open to some discussion, but if
such a system is required to transmit accurate movements of artists,
say, in a dramatic production, that which provides more accurate
Fig. 16.3. Coincidmt fi9cl.cof*. ht minrrplhoncs a! q l c
mdputfrom ~ i c A.
is r~~CIC"ICd
by PO Md
fm minaglronc B by QO
positional information is likely to be preferred. This therefore
suggests that the coincident microphone system should be used as a
basis. This, and its developments, will be discussed in the remainder
of the chapter.
In many programmes, particularly serious music, a simple set-up
employing one coincident microphone pair will be all that is necessary. For example, a symphony orchestra without a soloist, and
in some cases with a soloist, will usually be quite satisbctory with
this technique. Similarly, a small chamber group is also a simple
matter to reproduce in this way. Dficulties begin to arise, however, when, for example, the orchestra has a solo instrument and
the scoring is such that the solo line will tend to be obscured by the
orchestra. In monophonic balance technique, it would be simple
to add a second microphone close to the soloist and to fade this
up sufficiently to restore the desired balance. In stereophony,
however, although this can be done, some consideration of the
consequences of using various types of microphone is n e e s a y so
that the required result can be obtained.
We have seen that our main stereophonic microphone pair can
consist of two directional microphones with their axes at go0; in
this case the acceptance angle of the " stereo microphone " might
be expected to be goo. In fact the angle with figure-of-eight
microphones set a t go0 is rather less than this, about 80•‹,and with
cardioids rather more, as much as 160". Any sound which lies on
the left-hand extreme of this acceptance angle will appear to come
from the loudspeaker on the left and similarly sound on the other
extreme will appear to come from the right-hand side. This
result will be obtained, no matter how Eu away the sounds are
from the microphone. This fixed angle of acceptance, which
depends on the type of microphone pair used, can have a number
of disadvantages. For example, with a simple " single-microphone " balance of an orchestra there will be one position of the
microphone where the orchestra occupies the 111 width of the
reproduced sound stage. If the microphone is placed nearer than
this some instruments will be in the out-of-phase areas of the miphone polar characteristic (Fig. 16.3)~and if the microphone is
further away the orchestra will only take up part of the available
sound stage. Thus there may be difficulty in reconciling a suitable
width of reproduced sound with a satisfactory directlindirect sound
ratio. Furthermore, if a soloist is placed very close to a stereophonic microphone, the effect would be that the soloist would be
reproduced having an exaggerated width (Fig. 16.4). This is
obviously not required in normal circumstances, but may be of use
for a special dramatic effect.
Plale I 6.1. BBC ex)er~mental stereophon~cstudio control desk
Above: Plate 16.2. Neumann S M z stereophonic microphone. Bclow: Plate 16.3.
AKG (2.24 stereophonic microphone
Fortunately it is possible, by electrical means, to mod* the scale
of width of the reproduced sound. To understand how this can
be done, let us for a moment consider what would happen if we were
to connect the two channels of one stereo microphone together. In
this case, whatever information was in the left-hand channel will
now also appear in the right-hand channel and vice-versa, so that
both channels will have identical inputs. Our stereo " picture "
will have collapsed to a point source half-way between the two
Fig. 16.5. Simple width control-namwing
loudspeakers. If now, instead of " shorting together " the two
channels in this way, we connect a variable attenuator between them,
it should be possible by altering the value of this attenuator to
achieve any picture width from the full distance between the loudspeakers at maximum attenuation, to a point source in the centre
at minimum (Fig. I 6.5).
The foregoing statements are only true when both channels are
" in phase
Indeed, the supposition that both channels are in
phase is necessary to all the arguments so far in this chapter. If
the channels were not in phase, no centre image would be possible
and an unpleasant effect would be produced. To return to our
width control, if instead of connecting our attenuator so that it
" shorts " the channels in phase, we were to reverse the connections
to one channel, so that as the attenuator was varied the two channels
were shorted out of phase, we would find that a certain amount of
widening of the stereo picture is possible. I t is obviously not
possible to widen it to a very great extent, because when the channels are shorted out completely in the out-of-phase condition the
stereo effect disappears. The limited amount of widening that is
possible, however, can prove useful when the stereo microphone
has to be placed at such a distance from the instrumentalists that
their image in the sound picture is rather small. Such an occasion
might arise in a studio with a rather dead acoustic where, in o r d a
to achieve a satisfactory ratio of direct to indirect sound, the
microphone distance might need to be rather large.
In practice, it is technically difficult to make a control which can
be used for both widening and narrowing the sound picture when
it is necessary to have a phase switch at some position in the travel
of the control. Fortunately, another solution is possible. In the
narrowing case, one might consider that shorting channels "in
phase " is the same as adding them, and in the widening case,
shorting the channels " out of phase " is the same as subtracting one
from the other. Hence one might say that widening and narrowing
can be achieved by varying the amount of sum or difference signal
present in the two channels. It is a simple matter using transformers with double wound secondary windings to produce the sum
Fig. I 6.6. Sum and &$wna circuit
signal and the difference signal from our two stereo channels
(Fig. 16.6). A suitable control can then be devised which will
vary these two signals---our width control. The sum and difference
signals can then be combined with transformers as before in order
to extract the two channel signals again. Putting this another
way, if A and B are left- and right-hand signals:A+B=sum
- B = difference
(" M
" signal in international terminology)
(" S " signal in international terminology)
These are our two signals to vary with the width control (Fig. 16.7).
On recombining and taking the average:-
A secondary effect occurs during the operation of .the width
control, which will also affect the ratio of direct to indirect sound.
Fig. 16.7. W& rmtml in sum and d'f'n'@
As the picture width is narrowed, less reverberation will be reprod u d , and as it is widened, reverberation will be increased. This
is explained by the fact that it can be shown that the " S " signal
contains most of the reverberant sound.*
SoCo pod Spot Miaophones
From the above it will be evident that for a normal solo or spot
microphone a stereophonic microphone is not suitable. The solution to the problem is to use a monophonic microphone for the
soloist and to connect its output into both channels in such a way
that the image h m this microphone coincides with the weak image
of the soloist on the main microphone. The device used for this
operation is known as a " panoramic potentiometer " (Fig. 16.8),
the name arising from the fact that by its use a sound source can be
moved at will to any position in the width of the sound picture. In
the use of a spot microphone, care must be taken to see that there is
no confusion between the images due to this microphone and the
main microphone. The setting-up procedure is as follows: the
solo microphone channel gain control is advanced to such a point
Scc &an
an M S " Srersophony at the end of this chapter
that the effect of the spot microphone can be heard, the " pan. pot."
is moved until the image fiom this microphone coincides with that
h m the main microphone, and the gain control is adjusted to give
the correct balance. A check on whether the two image. have
been correctly overlaid is made by W g quickly fiom the stenophonic microphone to the solo spot microphone. Thm should,
of course, be no movement of the image. This principle can be
Fig. 16.8. Panoramic potenriomrtn-"pan. P6(."
extended to an almost infinite number of monophonic or spot
microphones and, in fact, in dance band and light music balances
where multi-microphone technique has become a part of the sound
produced, the use of a number of spot microphones is essential to
produce the required effect. It should be stressed that at all
times the stereophonic main microphone is essential in order to produce an overall reverberation or acoustic in which the sound can be
It would obviously be possible, indeed it has been done, to use
a number of spot microphones, each producing a point image of
an instrument or group of instruments, spaced across the sound
picture by means of " pan. pots.", and dispense with the stereophonic microphone altogether. This, however, does not produce
such a convincing result, since no sound appears in the space
between the instruments, whereas, of course, in the studio, this
space would be occupied by reverberation. A more convincing
reproduction is achieved, therefore, if the stereo microphone is used
as a basis and the spot microphones used for reinforcement only.
A further complication must be avoided. It is possible for any
two spot microphones to act as a spaced pair of stereophonic
microphones and thus produce spurious images of sounds within
their pick-up area. Care must be taken in siting spot microphones
to reduce this effect to a minimum, or blumng of the reproduced
sound stage will result.
16.8.2. Use of More Than One Stereophonic Microphone
Whilst the foregoing argument applies strictly to solo instruments
which are small in physical size, for example, solo violin and solo
flute, there may be other occasions when the solo instrument is
rather larger, like a piano, or indeed is not a soloist at all but a
chorus, accompanied by an orchestra. This chorus might not be
larger than the orchestra itself, but comparable in size, and placed to
one side or the other. A monophonic microphone in front of such
a chorus would produce only a point source of sound, and clearly
this is not what is required. In this case a stereo chorus microphone
is necessary in order to give width to the chorus and some means
must be found of placing the image of the chorus due to its microphone in the same place as the image on the main microphone, and
also of ensuring that the width of the chorus is again the same as the
width h m the main microphone. Fortunately this can be done.
Some side movement of the stereo microphone output can be
achieved by actual turning of the microphone. In addition, to a
limited extent a movement similar to the " pan. pot." in the case of
the monophonic microphone can be obtained by altering the relative gain of the left and right halves of the microphone. This can
be termed " off-set
We have already seen how the width control
can be used to adjust the width of chorus in the complete montage.
In any studio control equipment used for music or drama some
means of adding artificial reverberation is necessary, and stereophonic equipment is no exception. In this case, as well as the usual
mixture switch, some other facilities can be provided. Let us
assume that the reverberation is provided by a room. The reverberation room loudspeaker need not be paired since one loudspeaker
is all that is necessary to excite reverberation in the room. The
microphone, however, must be a stereophonic one, since we may
wish to create reverberation over the whole sound picture.
The feed for the loudspeaker can be taken from four different
sources:-the left-hand channel, the right-hand channel, the sum,
or the difference of the two. The return from the reverberation
mom microphone can also be made in several ways:-to the lefthand channel alone, the right-hand channel alone, stereo, the sum
of the two to both, or even to a " pan. POL" This last will produce
reverberation from any part of the reproduced sound stage. By
combining thesc functions in different ways, many effects can be
produced. For instance, in light music, if the strings arc on the
left-hand side of the orchestra, the left-hand channel can be fed
to the reverberation room, and the string reverberation thus pmduced returned to the studio control desk as stem, when it will be
heard over the whole of the sound picture. If it were returned on
the left-hand channel only, the reverberation would appear as a
point source at the extreme left of the picture. In a dramatic
production, it is possible to make someone walk from the left-hand
side to the right-hand side, being in a reverberant acoustic on the
left and a dead one on the right. It is also possible, by feeding the
reverberation room loudspeaker from the difference signal, to make
him walk from, apparently, a reverberant corridor on the left-hand
side of the picture through the centre of the stage in a dead acoustic
and out through a similar corridor on the right.
On a studio desk designed for stereo operation, it will by now be
evident that there will be a large number of individual controls
(Plate 16.1). For each stereo channel there will be a fader, a width
control, a reverberation mixture switch, a reverberation source
selector, an off-set control for moving the whole of the stereo picture
to right or.left, and one other control. This last is a pre-set control
used during the setting up process for each studo session to ensure
that the two halves of each microphone chain are perfectly balanced,
including of course the stereophonic microphone itself, the most
likely source of unbalance. Each spot microphone channel will
have its fader, a " pan. pot." and a reverberation mixture switch.
Facilities will also be provided for the playing in of stereo effects,
probably by tape, and for connecting monophonic disc recorded
effects to one or more spot channels.
Monitoring is, of course, necessary, and a P.P.M. having two
pointers, one for each channel, has been developed. Loudspeaker
feeds must, of course, be provided with suitable arrangements for
controlling their volume simultaneously. Since it is likely that the
" M " or sum signal of the two channels will be used as the signal
for the monophonic listener to a stereophonic programme, a switch
must be provided to feed this signal to one loudspeaker in order that
it may be checked.
This " M " signal should contain all the information necessary for
the mono~honiclistener. but care must be taken that the sound
produced $ as close as h b l e to that from a normal monophonic
2!5 I
balance. This necessary condition may be difficult with some
programme material.
Probably the most convenient method of recording twin-channel
stereophony is by means of magnetic tape. In this case, the two
tracks can be laid side by side, each track occupying half of the
tape, as is favoured in professional practice. Alternatively, as in
some domestic recorders, each track may occupy one quarter of the
tape, four tracks beiig used-two in each direction to save tapeeffectively extending the " half track " system to stereo (Fig. 16.9).
Few problems exist in the tape recording of stereo, though it is of
course essential that both recording and reproducing amplifiers in
the stereo machine be as identical as possible. In the early days of
stereo recording, the two half-track heads were placed one after the
other on the recording machine with the result that one track was
displaced by a distance depending on the head spacing, in relation
Fig. 16.9. (a) ) truck ~lnmpho& tape rrtmding. ( b ) ) track
stmopltmris tape recordins tracks r and 3 in rmc dir&
of to&
to the othex. This had the disadvantage that the head spacing
could well be different with different makes of machine, and the
more modern practice of stacking the two heads one above the
other is much more satisfactory, and has now become standard.
Standardisation has also been introduced as to which track is
associated with which channel. With the tape playing from left to
right with the magnetic coating away from the observer, the top
track should be associated with the left channel. This applies
with both half- and quarter-track trecording.
Whilst magnetic tape recording is undoubtedly the most satisfactory method for master recording and broadcasting work,
it is only slowly being accepted in the home and stereophonic disc
recording on a fine-groove long-playing disc is becoming increasingly
popular as a means of domestic reproduction of pre-recorded stereo
sound. In this case the two tracks are both recorded in the same
grooveon the dis~~effectively
one in eachwall ofthe groove, eachtrack
being cut at 45" to the plane of the disc. A complex cutter has been
designed to perform this h c t i o n , and the cutter head is fed in such
a manner that with two signals in phase, a lateral cut is produced
Fig. 16.10. Section though grwve of sbrcophonic disc rccmding
on the record and with the signals out of phase, a vertical (see
Fig. 16.10). Reproducing pick-ups likewise are designed to respond to the two sets of modulation and to produce two distinct
outputs, one for each channel. Great care must be exercised in the
design of both recording and reproducing heads to ensure that the
cross-talk is kept at a minimum. Because of the complex nature
of the recorded groove the reproducing stylus tip needs to have a
rather smaller radius than that commonly used for monophonic
long-playing records. This smaller tip radius necessitates a lighter
playing weight for the pick-up if undue harm is not to be done to
the surface of the record, since the pressure on the disc increases
as the radius decreases, according to a square law.
The radio transmission of stereophony has presented more
difficult problems than has the recording of the stereo signal. An
important requirement is that if a stereo programme is transmitted
by radio, a monophonic listener tuneci to the same transmission
must be able to hear the programme in a normal manner, i.e. his
reception should not be impaired by the fact that the programme is
also being broadcast stereophonically. This " compatibility ",
whilst obviously essential in a radio service, proves to be. difficult
where transmission systems are concerned since, in achieving it,
other undesirable factors become evident.
The simplest method of transmitting stereophony by radio, and
undoubtedly the most efficient in terms of pure stereophonic reproduction, so far as only one service area is concerned, is by using one
complete transmission chain for each channel. Obviously this is
highly inefficient in terms of economics, and in the use of a limited
number of transmission channels. Moreover, under these conditions neither transmitter is radiating an acceptable monophonic
signal, i.e. the system is not compatible. The experiments in
stereophonic broadcasting so far camed out by the BBC have been
using this system, but work is going on to develop a compatible
system which will enable the two stereophonic channels to be
carried by one radio transmitter, using only one V.H.F. channel.
Various systems for achieving this have been described, but so
far all of them have disadvantages of one form or another, and have
not been really suitable. Further development is necessary. A
brief description of some of these systems now follows.
16.7.1. E.M.I.
Pcrcival * System
The system proposed by E.M.I. is an ingenious one which would
enable stereophonic transmission to be carried not only by one
transmitter, but also by one land line, an important point.when the
type of programme distribution used in the United Kingdom is
considered. In this case, no matching of two land lines, one per
channel, would be necessary, thus saving considerable expense and
technical difficultv. The svstem works brieflv as follows: the two
channels are addkd togeth& and their sum * 'transmitted by the
radio system in the normal way as a monophonic signal. Experiment has shown that this sum signal produces, in most cases, an
acceptable signal for the monophonic listener. In addition to the
sum signal, a further signal derived from the two channels, termed
the " directional " signal, is also transmitted along the land line
and by the transmitter. This is possible because the directional
information supplied by the system can be compressed into a bandwidth of less than too CIS.and this can convenientlv be transmitted
at the upper end of &.audible spectrum withoit impairing the
programme material. At the receiving end, the sum signal is applied
to both channels and the directional information is extracted and
applied to the two channel amplifiers via electrical networks in such
a way that it controls the relative gain of the two amplifiers. Thus
it will be seen that with the left-hand amplifier a t zero gain and
the right-hand amplifier at maximum, the signal will appear to
come from the right of the picture, and similarly with the right-hand
amplifier at zero gain and that on the left at fuU gain the sound will
appear on the left. At gain settings between these limits sounds
can be made to appear at any point in the sound picture. The
supposition is that there is a " persistence of hearing " akin to the
well-known phenomenon of " persistence of vision ", and that
providing that the directional information acts quickly enough, a
complete sound picture will be built up. In effect, the directional
information acts like a high speed " pan. pot.", moving the sound
almost instantaneously to its correct location. Unfortunately, as
was described earlier in the section on microphone technique, the
end result is in some ways similar to the use of a number of " pan.
pots." with no overall stereo microphone, i.e. there is a lack of an
enveloping acoustic. Various other limitations are evident in the
system at present, but research and development are still proceeding.
16.7.2. The Crosby System
The system, proposed by the Crosby Corporation in the USA.,
is intended primarily for use with a V.H.F. radio transmitter and
again the sum of the two channels is transmitted in the normal
manner of a monophonic programme. This system can therefore
be considered to be compatible. In this case, however, no directional signal in the Percival sense is derived, but the difference signal
(see argument on sum and difference above) is transmitted on a
subcarrier along with the main transmission. This sub-carrier can
be of the order of 50 kc/s and the difference signal is frequency
modulated upon this. At the receiving end electrical networks
separate the sum and difference signals, pass them through a further
sum-and-difference or " matrixing " network, and produce the left
and right signals again. This system is capable of producing good
stereophonic reproduction, and indeed has been used for stereophonic experiments in the United States and in Holland.
Unfortunately, however, because the total bandwidth of the transmission is restricted the radiated sum signalis at a lower level than it
would be if it were radiated without the sub-carrier signal. Consequently the signal-to-noise ratio at the receiver is worse, by some
decibels, than in the monophonic case. This will, of c o w ,
affect the monophonic listener. The stereophonic listener will
also be affected in this respect since the sub-carrier channel has an
even worse signal-to-noise ratio than the main channel. The
matrixing network averages this noise out over both channels, but
the overall effect can be anything up to 2 0 dB worse than the
normal monophonic transmission. This system can work very well
within the primary service area of the transmitter, as is the case
with most American local broadcasting stations, but the signal-tonoise problem becoms acute when the high-powered broadcasting
system with large fringe areas, such as that used in the United
Kingdom, is attempted. Another difficulty with this system is that
the co-channel and adjacent channel interference may be worse
due to the presence of the sub-carrier.
The Mnllsud System
The system proposed by the Mullad Company also enables the
two channels to be caRied on one V.H.F. transmitter. In this case
the transmitter modulation input equipment is switched at a very
high speed, some .32,000 times per second, from one channel to the
other so that left and right channels are transmitted alternately
by the one transmitter. The monophonic listener has no means of
separa'ting these and so hears effectively the sum of the two channels,
a compatible signal. The stereophonic listener has an electronic
switch synchronised with the switch at the transmitter, and is thus
able to separate the two channels and feed them to their respective
loudspeakers. Again the stereophonic effect is well transmitted but
signal-to-noise and interference problems are still present.
United States System
The Federal Communications Commission (F.C.C.) of the United
States has now approved a system for stereophonic broadcasting by
stations in that country. This system is practically identical to
those proposed by the Zenith and General Electric Companies of
the U.S.A. and is a multiplex system for radiation by VHF, FM
The main channel canies the " compatible " sum or " M "
signal, and a sub-carrier, the difference or " S " signal. Unlike
the Crosby system described above the sub-camer is mnplitul
modulated, and the sub-carrier itself is s u p p d , only the sidebands being radiated. A second pilot sub-carrier, unmoddatcd,
is radiated at low level to ensure locking of the receiver circuits.
The " stereophonic " sub-carrier is at 38 kc/s, and the pilot subcarrier at 19 kc/s.
In addition to the stereophonic sub-carriers, the system allows
for the inclusion of one or more other frequency modulated subcarriers, designed for " storecasting
Whilst such a system has been found suitable for broadcasting
fiom small local stations, as in the U.S.A., it may be less suitable for
the system of nation-wide coverage used in the United Kingdom and
some other European countries. The problems of signal-to-noise
ratio, and interference between neighbouring channels still exist.
Whilst two ribbon microphones of the PGS or 4038 type can be
used for co-incident microphone stereophony, it is difficult to place
them close enough together because of their physical size, and to
maintain a balance of frequency response between them over the
required spectrum. Furthermore, the figureaf-eight characteristic
does not always produce the best possible effect, and so the polar
response should preferably be adjustable. Special microphones
have been developed for use in stereophony,'rnainly of the electrostatic type where the relative frequency response of the two capsules
can be controlled more accurately; and their small size enables two
to be mounted very close together. Two microphone capsules
and two microphone amplifiers can be conveniently mounted in
one case very little larger than the equivalent monophonic microphone.
Nmmann SM.2 (Plate 16.2)
This is a small stereophonic microphone using two capsules, each
similar to that used in the KM56 monophonic microphone. Provision is made for rotating one capsule in relation to the other to
enable them to be set at the required angle. Variable polar diagram control is provided for both capsules by means of separate
controls on the power unit.
16.8.2. AKG C q Microphone (Plate 16.3)
As might be guessed, this is in fact a combination in one case of
two C. I 2 microphones as described in Chapter 5. Again provision
is made for rotating one microphone capsule in relation to the other
and the complete range of polar characteristics available with the
C. I 2 is present with each capsule.
An alternative system of microphone technique has also been
" M " and " S'signals directly, without
the necessity of sum and difference networks. In this case the two
microphones have different polar characteristics, as shown in Fig.
used, which will give the
Fig. 16.I I .
MIS m i c m f ~ hqstun
16.I I . One is a forward facing cardioid or an omni-directional
microphone, and the other a figure-of-eight with its live axis a t go0
to the general direction of the sound source.
The forward-Gcing microphone provides the " M " signal, and
the sideways figure-of-eight the " S " signal. It will be seen
immediately that little direct sound should reach the " S "
microphone, almost the entire output of this microphone being
due to the reverberant sound in the studio.*
I t has been suggested that since this arrangement generates the
" M " and " S " signals directly a saving of equipment in a transcan result. Unfortunately it is still necessary to
mission syste~~
extract the A and B signals for monitoring purposes so this saving
seems to be an illusion.
The effectiveness of the stereophonic or derived monophonic
reproduction by this system is no different from that of the A and B
system described earlier.
Hcmcc it beanner apparent that the inor dwease of the " S " si&
will alter the revertmaion as wcll as the width of the picture ( s e c t i o n 16.3).
In Chapter 2 a description is given of the use of the den'bel to
compare sound intensities This unit is also used to denote electrical
levels in the programme chain, the basis of comparison beiig the
standard level of I milliwatt (0-001 watt) in a 600 ohm circuit.
This is known as zero level, and other levels above or below this are
quoted as
x dB or
x dB, the statement " above (or below)
zero level" being taken for granted. Rogramme at about this
level gives comfortable listening on headphones or telephoneshence its use in ring main circuits, and as the studio sending level.
A level of 0.1 watts, for example, would be quoted as 20 dB,
since 10 loao -= 10 loglO 100 = 20
The gain of amplifiers, and the &c.s of attmuators arc also expressed
in dB, the actual number of dB being calculated from the formula
number of dB =
loglo Pl
where PIand P, are the input and output powers.
In most studio equipment, the v o I ~ elevel is more important
than the pmucr I
.To express the voltage gains etc. in dB a
modified form of the formula is used.
If P
is the power ratio,
Vl*where Vlis the voltage and
P, = -
<the impedance in the
Also, P8 = 5
Hence number of (voltage) dB = 20 log VI
This shows that the number of voltage dB is twice the power dB
for a given ratio, and the reason is that the power ratio in a given
circuit is proportional to the square of the voltage ratio. (This is
only strictly true if the impedances at the two points are equal.)
(i) How many dB separate 500 watts and 5 watts?
Number of dB =
10 log,,
500 = 10 log,,
loo = 20 dB.
(ii) What is the gain of rn amplifier which gives an output of
8 volts for an input of 0.0008 volts?
Number of (voltage) dB = 20 log,,
(iii) How many watts of electrical power correspond to
with reference to zero level?
40 = 10 log,,
'00 I
.: log,,
P =4
P = 104
= 10 watts
+ 40 dB
Table Ax.
If we let a point P move round a circle at a uniform speed, the
angle " a " will take up all values from o0 to 360•‹, and y will rise
to a maximum in one direction, fall to zero, reach maximum in the
Fis. A. I .
l7u sine-warn graph of Sin 13against 0
opposite direction, and again fall to zero. Drawing a graph,
plottingy against the angle 9, we obtain a smooth curve (Fig. A. I).
Two features should be noted:(a) At all positions of P, OP = r, the radius of the circle, and if we
let r = I, we see that plotting 9 againsty is equivalent to plotting
9 against sin 9. This curve is therefore called a sine wave.
@) As P moves round the circle with a uniform velocity, M moves
to and fro, imitating the bob of a pendulum or the prong of a
tuning-fork. This motion is called Simple Harmonic Motion, and
is a characteristic of a pure tone. Thus all the particles of a
tuning-fork take up such a vibration. This is communicated to
all the air particles in the vicinity, and to the eardrums of an
The distance of a vibrating particle from its mean position x at
any instant is called its dirplacement.
The maximum displacement is called the amplitudosymbol a.
A complete vibration of N through all positions and back again is
called I cycle-and is related to I circle of revolution of P.
The stage which a particle has reached in its cycle is called its
phuse. Thus crests are in phase, and a crest and a trough are out
of phase by 180".
Note: The horizontal scale in Fig. A.I might equally well be a
time scale.
RAYLEIOH, LORD. l 7 u q of Sound, Macmillan (1896).
PLETCHER, H., Sptvch and H&g
in Cavnwticotwn, D. Van
Nostrand Company
Inc. (1953).
CAINN, H. A, and EISENBERO, P., Tonal range and intmcilyprefc~mes, Pm.1.R.E.
(Sept. 1945)BOMERVIUE, T., and BROWNLEIS, 9. P.,
k d f i ~ f f f m e c ~ , BBc
(Jan. 1949).
JBANS, J., StiCNC and Music, C.U.P. (1938) Paperback (1961).
WOOD, A, l
h Plysics OfMusic, Methuen & Co. Ltd. 4th ed. ( 1 ~ 7 ) .
OWN, H. P., M
d Eqincning, McGraw-Hill Publishing Co. Ltd. (1952).
KNUWBN, V. o., Archircctwal IqCmtics, John Wiley & Sons Ltd. (1932).
M d , IlXe Books Ltd. (1955).
BBC T 6 c W Imbucthn S3.
BERRY, s. D.
AmpliJFrrs fm Ahmi B n w h d n g , BBC Enginwring
DbiShtl Moro~gr~$Ih
NO. 26 ( h g . 1959).
NOS, s. w., Principles of Trmrrizim C+,
2nd ed., Iliffe h k a Ltd. (1961).
TERMAN, P. E., Eh&&
and Rndio Engineering, 4th ed., McGraw-Hill
h d . (1.955).
P., Ra&o Design& H d o o k , 4th ed., Iliffe h k a Ltd.
Engin#ring Training M a d ,niffe Booka Ltd. (1952).
SHORTER, D. E. I, and RARWOOD, H. D., 7 7 Duign
of a Ribh-IIPs Ra~m
Godiolr MkqblroM fa Bro&asi TI&,
BBC Engincaing Momgr4ph
N o - 40 (k
. D., Ih
Kwh f&& Conrmrrdo&f'~ Mirwruiriar b Ambht Noiu,BBC Engiffcning D
d Momgraph No. 7 (June
sluvrnrorn, EL J. VON, and WEBER, w., Cbnthw a d n o p h , H&h Frqwnz
Elrkhwkuslik., 46, p. 187-92 (1935).STUDIO ~ Y E N TWB
A, BBC Tkhnicd Insbl(CtMD S5.
PROCRAVMB ~ETERS, BBC Eqbmrhg Training Suppkmsnt No. 6.
~ O R T E R , D. E. L, A
O f p n f ~ r m r m c acritaia and dacign wnridcreiiaufw hifh
[email protected] l o u d r ~ sJ, . Inst. Ebc. Engrs. ((London), Pt.B. (Nov. 1958).
. 8.) BBC
Enginosring Dioiria Monograph No. 5 (Feb. 1956).
BERNHART, J., Trait4 & Prirs & Son, Editions Eyrollea (1949).
LON0 W A M AND MEDIUM W A M PROPAOATION, BBC Engincsrind Tfaislgplmvnt No. 5.
BBNNMGTON, T. w., 5htumw Radio and the Ionogrhar, 2nd d,Iliffe Boob
Ltd. (1950).
~ P B Q ~ C
g S y P p No.
~ g.
B.P. 3 9 4 3 ~ 5 .
M., DUTPON, O. F., and
[email protected]~Itrrn,3.
A. D.,
lh S&[email protected]
Engrs. (London), lo4 k B . (1957).
I N . Elm. Engrs. (Lo&), PrB. Suppl. No. 14 (1959).
M., F w h U~&
On [email protected] d w, W h ? h World
(April & M a y 1960).
.I and PO. J., Swnwq? of the ?+ant Pasition ofSbrrP
p k Blwdccrsting, BBC Enginuting Dioicion Mmgr(Iph No.29 (April I 960).
,., O @ d h d h C h On [email protected] mrd Sludio T
in Stncophony, BBC Engineering Dinicion Monograph No. 38 (Scpt. 1961).
BBC Instructions and Training Suppluncne an written primarily for the
use of the BBC staff, but a limited number of w p i a is available and may
be obtained at reasonable cost on application to the Editor, Technical
Instructions, Broadcasting H o w , London, W. I.
cavity 43
membrane 41
Absorption, sound 36
Absorption coefficient 40
Absorption methods 40
Acoustic effects reproduction 234
Acoustic studio screens I 5, 200
Acoustics, studio 34-37
Air-borne noise 34
Air column
excitation 25
musical 24
AKG Cnq stereophonic microphone
Akustische und Kinoger'dte G.m.b.H.,
electrostatic microphone 89
Alternating current 59
and capacitance 61
and inductance 61
and resistance 60
and transformer 61
Ampere 49
for reverberation plate I 10
gain of 259
microphone 62, I I 7
outside broadcast 141
standby 122, 129
Amplitude 8,262
modulation 1 88
Anode 64
Antinodes 13
Apparent loudness 8 .
Artificial reverberation 108,22 I,235
stereophonic 249
Atom 48
Attenuation distortion 100,1g1,192-g3
Attenuators 235
loss of 259
output 141
variable 104
Auditory canal 16
Autetransforma 58
Bacon, Francis, New Atlantis 237
BatBe, loudspeaker gq
Balance, &t
of acoustic anditions
I I2
Balanced circuit 104
Balancing I H
Basilar membrane I 7
Bass clarinet 29
Bass correction unit 202
Bassoon 29
Beat frequency 6
Be1 19
Bell Telephone Laboratories 238, nqo
Bias 65
Bias oscillator I 56
Biscuit 75, 84
Blurnlein, A. D. 239
Blurnlein stereophonic system 242
Boom operator 46
Boominess 197
Brass bands, microphone placinp:
- for
Brass instruments, microphone placing
for 207
Broadcasting chain and distortion
Buzzers 57
Capacitance 54
alternating current and 61
Capacitor microphone 81
Capacitors 54, 61
Carbon microphone 79, 83
Cardioid microphone 76-78, 199,2 14,
radio frequency 188
Cathode 64
Cavity absorbm 43
'Cello. See Violoncello
Chamber music, microphone placing
for 212
Channel faders, independent 132
Channel switching 106
Cherry and Leakey 239
Choir, microphone placing for 202
Choir with orchestra, microphone
placing for 2I 8
balanced 104
equalisiig 193
filter 63
public address r 53
tuned 63
Clarinet ng
placing for 2 I 3
Clavichord, microphone placing for 207
Clean feed I I 3. 124. 125, 130, I 34, 135
Clean-feed and 'ZTB 124
Clean-feed talk-back key I 26
Cochlea 17
Combination t o n e 5
Compression 225
Concertos, microphone placing for I 2 7
Condenser 54
condense^ microphone 81
Consonance 18
Continuity suite I 81-82
Contra bas. Su Double b a s
Contra-bassoon 29
control cabinet, Type A I 15
Control desks. Scc Studio control
Control grid 65
Control room 182-83
Cor anglais 29
Cornet 30
Coupled system I I
Court-room scene 201
Covent Garden Opera House, BBC
equipment at 135
Crosby stereophonic system 254-55
Crystal microphone 82
types in current use gr
Cue circuit, master 134
Cue lights I I 2
selection 129
Current divider 52
Cutting stylus 169
Cycle 262
Dkc recarding d
Damping g
Dance bands, microphone placing for
Ear, human 16,226
Decibel 20, 259
Difference frequ&cy 6
Diffusion of sounds qq
Diode 63,65
Direction effects of loudspeaker 94
Directional hearing 239
Directivity (of sound source) I I
Disc recording and repduction 16879
charaaeristio, 169
coarse groove I 70
direct 17I
processing I 7 I
equalisation in I 73
fine-groove (microgroove) 170
h e g r o o v e reproducing d a k DRDI)
groove-locating unit GLU/gB I 75
optical grocwe indication I 79
pitch 170
pre-fade listening I 7+, I 79
reproducing desk I 77
prqce+9ing 171
qulck-start device, I $, I 78
recording head 1%
reproducing desk RPZII...I 79
reproducing head Type E.M.I. 12,
stampa I 72
stereophonic 252
s m e o q e 174,178
tranxnphon r e c o h g s 170
turntable dak, G 12 tilten on 137
turntable desk, TD,'7-.. I 72-77
turntable speeds I 70
D i i o n s , microphone placing for I 98
Displacement of \ibrating partide 262
Dissonant interval 19
Distortion 196, 228
attenuation 191. 1-3
frequency 192
in broadcasting chain 180-93
typs of 191
Distottion units 129
Domains 55
Double bas 26
microphone placing for 2%
Dramatic productiow, microphone
placing for 1gg-201
Dynamic microphone 79
Dynamic range 225
Eardrum 16
Echo 36, 108, 147
and feedback 235
Echo chain, Type A 121
Echo cut key 121
Echo mixture switch I 21
Echo selection 129
Echo source 130
Edge tones 2j, 27
Effects unit, portable 136
Eigentones 43
Electric bells 57
Electric current 49
Electrical interference 189
Electromagnet 55
Electromagnetic waves 185
Electromotive force (c.m.f.) 49
ElectroniEmusic 234
Electrostatic loudspeakers 6,
- . loo
constant change 96-97
Electrostatic microphones 77,78,81--82
2 ~ 6
ty$3 in current use 88
Elements 48
E.M.I. mametic
tape recorder TRlgo
E.M.I. midget recorder 84
Type L.2 ...165
E.M.I. " Percival " stereophonic system 253
Enclosure, loudspeaker 94
End correction 24, 25
English horn 29
Envelope 189
modulation 239
Equal temperament 19
Equalisation in disc recording 173
E q u a F r 183
Equallsing circuit 193
Equilibrium intensity 36
Erase head I 55. 157
Eustachian tube 16
Excitation of air column 25
Faders 104, 218
channel, independent I 32
constant-impedance I 06
group '09. 131
main control I 17
reverberation 109
Fading 187
Feedback, echo and 235
Fmograph tap recorder 162
Ficord miniature tape recorder, hiodd
Fidelity in reproduced sounds 190-93
F i m e n t 64
Filter circuits 63
G12 ...137
OW pasS 139
wx for sound effects 2 3 h 7
Fie-groove reproducing desk DRDl5
Forced vibrations g
Formant 22, 33
F m vibrations g
French horn 2g
Frequcncics of ear 17
Frequency 32 59
fundamental 4
resonant 63
Frequency distortion 192
Frequency modulation 189
Frequency response
of microphone 7 I
modifications to 135
Frequency shift P.A. system 150
Fundamental frequency 4
Gain of amplifiers 259
Gramophone recording. See D
c recording
Green cue lights 134
Groove-locating unit GLU/gB 175
Ground wave 186, 188
Group fader 109, 131
Group selection I 29
Group switching 106
Groups of singers and instruments,
microphone placing for 209
Guitar, microphone placing for 208
Hair cells I 7
Harmonic structure 235
Harmonics 4, 5, 204
Harp 26
microphone placing for 205
Harpsichord, microphone placing for
Headphones I I I
Heater 64
H e w 55
Holes 66,67
Horn loudspeaker 149
Howl-back 35
Howl-round "3, 148
Human ear 16.226
Human voice j 2
Humidity, effect on velocity of sound 8
Hybrid transformers 58, I 08, I r 4, 12 1
Impedance 61,62
output 62
In phase 60, 1% 245
Inductance 5 5 5 7
alternating current and 61
Inductor 61
Instability of PA. system 148
Insulation, sound gq
Insulators 49
Intensity 8, 17, 19
Intderence 13
e ects, mixing 1%
Interviews, microphone placing for I g8
Ionosphere 187, 188
Jacks, plugs and 107
Jazz group, microphone placing for 224
Just intonation 19
Leakey and Chary 239
Leevers-Rich tape reproducer 163
Light, speed of 185
Light music ensembles, microphone
placing for 2 12
Limiter 228, 229
Line source units 154
Linearity 62
Lip microphone 132, 134
Listening levels loo, IOI
Long-playing records I 70
Long waveband I 86
Loops 13
Loss of attenuatom 259
Loudness 8, 19
P.A. systems 146
Loudspeaker baffle gq
Loudspeaker enclosure gq
Loudspeakem 59, 93-102, I I I
O 149 ~
correct use of I O I
directional effects of 94
electrostatic 96, loo
constant change 96-97
horn I 49
in current use g8
listening levels 100, 1 0 1
moving-coil 93
multi-unit 95
phasing of I 7
polarising voftage 96-98
public address systems 147, 149
reverberation room 249, 250
standby 123
Magnetism 55
" Magnetophon "
tape recorder I 55
Marwni Studio Console I 26
Medium waveband 186
Membrane absorbers 41
Microfarad 54
Microphone a m p l i m 62, I I 7
Microphone correction unit qq,63, I 35
Microphone distance 44
Microphone reflectors, parabolic I 5
Microphona 7 1-92
capacitor 81
carbon 79,83
cardioid 76-78, 199, 2 1 4 257
coincident 241
condenser 8 I
crystal 82
types in current use g1
curtain of 240
directional properties 74
dynamic 79
electrostatic 77, 78.81-82. 256
types in current use 88
figure-of-eight 242,256, 257
frequency response of 7 I
in current use 83
interference effects in mixing 194
lip 132, 134
monophon~c247, 248
moving-coil 59, 79
personal 85
v e s in c u m n t use 83
ornnl-directional 241, 257
output level of 71
placing 194-224
brass bands nrg
brass instruments 207
chamber music 2 12
choir 202
choir with orchestra 218
classical orchestra 2 13
close technique 196
concertos 2 I 7
dance bands 222
discussions 198
dramatic productions lgg-201
early keyboard instruments 207
group of singers and instruments
guitar 208
interviews 198
jazz group 224
light music ensembles 212
military bands 22 I
musical instruments 203
organ 208
percussion instruments 208
piano accordian 209
pianoforte 205-07
pianoforte (two) 21 I
singers (solo) 201
solo instruments 203
solo instruments with piano Pro
songs with orchestra 2 18
songs at the piano 21 I
songs with piano nlo
string instruments 203-05
symphony orchestra 214-17
Musical instruments
microphone placing for 203-94
groups 209
solo 210
theory of 22
wind 27-30
Musical scale 17
Musique concrtte 234
Nagra tape record-Model
I 66
rib&n 80, 81, 198, 214.
n o i s e - c a n d i g lip nbbon 87
types in current use 85
SO10 '247
spot 22 I, 222,242,247~tere~phonic
243-49, 25?t
MIS system 257
tape-racorder I 65
Type A I 17
variable polar diagram 78
Middle ear 16
Military bands, microphone placing
for 22 1
Mixer suite B. I. (Broadcasting H o w )
dean-feed facilities I 34
cue circuit, masta I 34
green cue lights I 34
independent channel faders I 32
Lip microphone 132
studio red lights 1%
talk-back lip microphone 134
talk-backlpre-We keys 133
Mixem, four-channel 182
bfking 104, 131
interference effects 1%
Modulation 188
amplitude 188
frequency 189
Modulation envelope 239
Monaural system q6
Monitoring 250
Monophonic system 46
Moving-coil loudspeaku. Scs Loudspeakers
Movinazoil microphone. Scs MicropKones
MIS stereophonic microphone system
Mullard stereophonic system 255
Music, volume of. Scs m
Musical amustics 1-3
Musical air columns 24
x I rB
Natural frequency g
Neumann SMn stereophonic microphone 256
New Atlmtis, by Francis Bacon 237
Nodes 13
Noise 7 I
air-borne 34
structure-borne 34
typs of 191
N-pn transistor 67
Nucleus 48
Oboe 28
Obstacle dfect 1472, 75.
Octave 4
Ohm 50
Ohm's law o
Open-air J e c t 200
Orchestras, microphone placing for
classical 2 I 3
Syrnph0ny 214-1 7
variety 224
Organ, microphone placing for 208
Oscillators 236
Ossicles 16
Out of phase r g5
Output attenuator 141
Output impedance 62
Output level of microphone 71
Outside broadcast equipment 140-43
amplifier I 4 I
self-operated 92, 145
Outside broadcast lines 143
tests on I 44
Outside broadcasts I 40-45
Outside sources I 26
Oval window 17
Panoramic potentiometer 247-.+8
Parabolic microphone reflectors I 5
Paris, BBC studio in I 35
Peak programme meter 11.1, 230,
931, 250
Percussion instrument3 30
microphone placing for 208
Phase 6, 262
See a h In phase; Out of phase
Phase distortion 191
Phasing of loudspeakers 147
Phon 20
Piano accordian, microphone placing
for 209
Pianoforte 31
microphone placing for 205+7
(two), microphone placing for 2 I I
Piccolo 28
Piezo-electric &ect 82
Pistol shots 227
Pitch 3, 4
Pitch (dimrecording) 170
Plane waves 12
Plate, reverberation I lo
Player's Lips 25
Plugs and jacks 107
P-n-p transistor 68
Pocket preamplifier 91, gn
Polar characteristic of microphone 71
Polar diagram 74, 77, 80
cardioid 78,89
figure-of-eight $5,8 I , 89
selection unit 89
Polarising voltage, loudspeakers 96-98
Polarity 55
Portable effects unit 136
Post office lines 183-85
Potential divider 52
Power 8,53, 60
P.P.M. I I I, 230, 231,250
Pre-amplifia, pocket g1,92
Pre-fade facilities I 29
Pre-fade listening I 74, I 79
operation, bass tip
UP m 73
Presto reproducing desk I 77
Programme junctions 230
Programme meter readings 230
Programme ring-main switch I I 2
Programme volume
compression 225
control of 2 2 5 3 1
dynamic range 225
limiter 228, 229
manual control 226,227
preferred volume of s p m h and
music 228
programme meter readings 230
range of sounds in music and speech
volume limits 225
Propagation of radio waves 185
Protons g
Public a%dras I I 4
Public address circuits 153
setf-contained I 53
Public address equipment
in current use 154
stereophonic I 51
Type B '53
Public address selection I29
Public address system
delayed I 50
frequency shift 150
instability of I 48
loudness 146
loudspeaken 149
phasing oi loudspmke~147
quality 147
sense of direction 148
Push-button selection 130
Quad quality control unit 138
Quality of public addsystems 147
Quick-start devices 1$5, I 78
Radiator 22
Radio frequencies 185
Radio frequency carrier 188
Radio transmission of stereophony
Radio waves, propagation of 185
Radiophonic effects 234-37
- (BBC)
. 234,237
Rayleigh, Lord, nLeov of Sound 239
Reactance 61
Recorded eKects 233
disc. Su Disc recording
stereophonic. Scs Stueophony
tape. See Tape
Recording head I 55
Rectifier 65
Red lights, shldio 134
Reeds 25
Reflected sounds 197
Reflection 14
Relays 57
Repeatem 185
Repetitive sounds 236
Reproduced sounds, f i d d i t ~in 190-93
Reproducing head 155
Type E.M.I.12 ...172
Resistance 50
alternating currents and 60
in parallel 53
in parallel 52, 53
in series 51
variable 52
Resonance g
sharpness of ro,63
Resonant frequency g, 63
Resonators 22
Reverberation 36, 200
artificial r 08, 22 I (See also Echo)
control 39
fader log
machines 109
plate I 10
room 109
loudspeaker 249, 250
stereophonic 249
time 37
optimum 38
studios, auditoria, etc. 39
Ribbon microphone 80,81, 198, 2 14
types in current use 85
Ring-main switch, programme I I 2
Room resonances 43
Root mean square value 60
Round window 17
Sabiie, W. C. 37
Scale, musical I 7
Screens, studio 15, 200
Self-operated outside broadcast equip
See Outside broadcast
Sense of direction of P.A. system 148
Short waveband 187
Simple .harmonic motion 3, 261
Simultaneous broadcast system r 83
Sine wave 261
Sine wave tones 236
Singers, microphone placing for
groups 209
solo 201
Skip distance 187
Sky wave 186, 187, 188
Solo instruments, microphone placing
for 203
Songs at the piano, microphone placing
for 211
Songs with orchestra, microphone
placing for 2 18
Songs with piano, microphone placing
for 210
theorv of 220
Sound absorption 36
Sound effects 168, 232-37
radiophonic 234-37
recorded 233
reproduction of 234
recording of 233
repetitive sounds 236
spot effects 232
tape effects 233
white noise generator 237
Sound insulation 34
Sound-proofing 34
Sound transmissions, fidelity in 190-93
Sound waves I , 2, 3
Sounds, diffusion of 4
Source selection, type B 128
Speech, volume of. See Programme
Speed of light 185
Spherical waves 1 I
Spot microphone. See Microphones
Square-wave shaping device 236
Standard musical pitch 3, 4
Standard Telephones and Cables Limited, movingcoil microphones 83
Standby amplifiers 122, 129
Standby loudspeaker r 23
Standing waves 13, 197
Static leak 140
Stereophony 46, 238-57
ancillary equipment 250
artificial reverberation 249
Blumlein system 242
coincident microphone system 242
&rectional hearing 239
domestic 243
history 238
microphone technique 243-49
microphones 256
monitoring 250
monophonic balance technique 243
M/S system 257
public address 151
radio transmission 252-56
Crosby y t e m 254-55
Percival " system 253-54
Mullard system 255
United States system 255-56
recording 239, 25 I -52
wavefront system 240
Storecasting 256
String instruments 25
microphone placing for 203-05
Strings 23
Stroboscope r 74, r 78
Structure-borne noise 34
Studio acoustics 34-47 Studio console, Marconi 126
Studio control desks and equipment
BBC studio in Paris 135
Covent Garden Opera House 135
mixer suite B. r (BroadcastingH o w )
Tape amtimud
stereophonic 251
Television studios 45
Tanpcrature effect on velocity of
sound 7
Tempered semitones I g
liuorp of S d , by Lord Rayleigh 239
T h m i o n i c valve 64
T* B 12d-30
Studio output, means of listening to r I I Threshold of feeling I 7
Threshold of hearing I 7
Studio red lights I 34
Timbre 5
Studio screens 15, 200
Top cut 94
Studio tables 197
Transformen 58
Supply cabinet, Type A r 14
alternating current and 61
Switch, programme ring-main I 12
auto 58
Switching, group and channel 106
hybrid 58, 108, 1 1 4 121
Symphony orchestra, microphone placmatching 79, 81
ing for 214-17
Transistors 66
n-pn 67
Talk-back I 12, I 18, I 19, 120, 121
P-n-P 68
Transmitting station I 85-90
Talk-back key 126
Triode 63,65,66, 68
Talk-backlprefade keys I 33
Triode amplifier 65
Talks, microphone placing for 196
Trombone 30
Tape effects 233
microphone placing for 207
Tape feedback 235
Trumpet 30
Tape recorders
microphone placing for 207
battery-operated 233
Tuba so
domestic 156
~ & e d * E i u i t s63
elements or I 55
E.M.I. magnetic recorder T R / 9 Tuning-fork 2, 3, 5
Turntable desks. See Disc recording
E.M.I. midget recorder-Type L.2 Tweeter 96
Two-way working I 24
Type A studio equipment I 14-26
erasing process I57
artificial reverberation 120
Ferrograph 162
control cabinet I I 5
Ficord miniature recorder-Model
control desk I 15
IA 165
echo chain I 2 I
Leevers-Rich taqf reproducer 163
faders, constant-impedance I r 7
" Magnetophon
individual microphone amplifiers r 17
microphones 165
main control I r 7
miniature battery-operated 165
Mark I1 I 16
Nagra Model I r I B 166
Mark V 116, 121
recording process 157
Mark VII I 16
remote control of I 63
microphone and group faders I r 7
reproducing head I 58
public address 153
reproducing roc- r 58
standby amplilim 122
tape speed I &
standby loudspeaker 123
tape transport mechanism 160
supply cabinet I r 4
Tape recording and reproduction 155Type B studio equipment 1 2 8 3 0
clean-feed facilities I 30
advantages and disadvantages 160f-i rcue light selection 129
distortion units 129
Broadcasting House arrangement
echo selection 129
group selection r 29
editme 161
pre-fade facilities I 29
equaLtion in 159
public address 153 local recording/reproduction I 64
public address selection I 29
sound effects 233-36
T y p e B studio equipment wnlinucd
push-button selection I 30
source selection 128
standby amplifiers 129
talks and special desks 130
United States stereophonic system
Valve, thennionic 64
Valve voltmeter I I I
Variable attenuaton 104
Variable correction unit 137
Variable resistors 52
Variety orchstras, microphone placing
for 224
Velocity of sound 6
humidity effect on 8
temperature effect on 7
V.H.F. 186, 189
forced g
free Q
strings 23
. -microphone placiig for 204
Violin 25
microphone plating for 203
Violoncello 26
microphone vlacirig for 204
Vocal group, microphone placimg for
Voice, human 32
Voltmeter, valve I I I
Volts 49
Volume, control of. Sss Programme
Volume indicator I 10
Volume limits 225
Volume unit meter I 10, 1I I
Watt 53
and their uses 187
long 186
medium 186
short 187
V.H.F. 186, 189
Waveform 5
Wavefront stereophonic system 240-42
Wavelength 6
White noise generator 237
Wind instrumenB 27
Scc also Brass; Woodwind
Widshields 84, 89, go
Woodwind instruments, microphone
placing for 207
Su also Wind instruments
Woofer 95
Zero level 259
Zen, programme
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF