Architectural

Acoustics
Briiel & Kjaer
ARCHITECTURAL A C O U STIC S

by

K.B.Ginn.M.Sc.

November 1 978
2nd edition 1st print

ISBN 87 87355 24 8
 CONTENTS

FOREW ORD .......................................................................................................... 7

INTR O D UC TIO N .................................................................................................. 8

CHAPTER 1. FUNDAMENTALS AND D EFINITIO NS ................................ 9

1.1. NATURE OF SOUND ..................................................................... 9

1.2. WAVE TERMINOLOGY ................................................................. 10
Plane wave ....................................................................... 10
Diverging wave ................................................................. 10
Spherical wave ................................................................. 11
Progressive wave ............................................................. 11
Standing wave .................................................................. 11
1.3. FREQUENCY OF SOUND ............................................................. 11
1.4. SPEED OF SOUND ....................................................................... 12
1.5. WAVELENGTH .............................................................................. 13
1.6. DISPLACEMENT............................................................................ 13
1.7. AMPLITUDE .................................................................................. 13
1.8. PARTICLE VELOCITY ................................................................... 14
1.9. SOUND PRESSURE ...................................................................... 14
1.10. SPECIFIC ACOUSTIC IMPEDANCE ............................................ 14
1 .11. DECIBEL ......................................................................................... 15
1.12. PEAK, AVERAGE AND RMS ...................................................... 15
1.13. ENERGY DENSITY ........................................................................ 16
1.14. INTENSITY ...................................................................................... 17
1.15. INTENSITY LEVEL ......................................................................... 17
1.16. SOUND PRESSURE LEVEL ......................................................... 18
1.17. SOUND POWER LEVEL ............................................................... 19
1.18. SOUND SOURCES: THEORETICAL AND PRACTICAL............ 20
1.19. MONOPOLE OR SIMPLE SOUND SOURCE ............................. 20
1.20. DIPOLE ........................................................................................... 22
1 .21. DIRECTIVITY PATTERN ................................................................. 24
1.22. DIRECTIVITY FACTOR AND DIRECTIVITY INDEX .................... 26
1.23. SOUND FIELD OF A SOUND SOURCE ..................................... 27
1.24. THE EAR ......................................................................................... 29
1.25. LOUDNESS, PHONS AND SONES ............................................ 31
1.26. SOUND LEVEL METERS AND WEIGHTING NETWORKS ...... 33
CHAPTER 2. ACOUSTICS OF ROOMS ........................................................ 34

2 .1 . DEFINITION OF ROOM ACOUSTICS ....................................... 34

2 .2. GEOMETRICAL ROOM ACOUSTICS ....................................... 34
2 .3 . GROWTH AND DECAY OF SOUND IN A ROOM ................... 36
2.4. REVERBERATION TIME .............................................................. 37
2 .5. ABSORPTION COEFFICIENT ..................................................... 37
2.6. DERIVATION OF FORMULAE FOR
REVERBERATION TIME ............................................................... 38
2.7. WAVE THEORY OF ROOM ACOUSTICS ................................. 39
2.8. PRINCIPLES FOR DESIGN OF ROOMS AND AUDITORIA .... 46
2.9. DESIGN OF ROOMS FOR SPEECH .......................................... 48
2 .1 0 . DESIGN OF ROOMS FOR MUSIC ............................................ 51
Loudness .......................................................................... 51
Reverberation ................................................................... 52
D efinition .......................................................................... 52
Fullness of tone ...:.......................................................... 52
No obvious faults ............................................................. 52
Intimacy (or presence)..................................................... 52
M usicians' criteria ........................................................... 53
2 .11 . REFLECTORS, ABSORBERS AND RESONATORS ................. 53
Sound reflectors ............................................................. 53
Sound absorbers ............................................................. 53
Panel sound absorbers .................................................. 55
Resonator absorbers ...................................................... 56
Perforated panel absorbers ........................................... 58
Functional absorbers ..................................................... 60

CHAPTER 3. ACOUSTICS OF BUILDINGS .............................................. 61

3.1. INTRODUCTION ........................................................................... 61

3.2. SOUND GENERATION MECHANISMS ..................................... 61
3.3. SOUND INSULATION ................................................................... 62
3.4. AIRBORNE SOUND INSULATION ............................................. 62
3.5. IMPACT SOUND INSULATION .................................................. 65
3.6. AIRBORNE SOUND REDUCTION INDEX FOR A
SOLID, HOMOGENEOUS, IMPERVIOUS WALL ..................... 67
3 .7. COINCIDENCE EFFECT ................................................................ 68
3.8. METHODS OF IMPROVING AIRBORNE SOUND
INSULATION OF BUILDING ELEMENTS ................................... 70
Damping ................................... ........................................ 70
Double leafed elements ................................................. 70
Flanking transmission ..................................................... 72
Doors ................................................................................. 73
Outer w alls and w indow s ............................................. 74
 Floor-ceiling elements ..................................................... 74
Floating floors .................................................................... 75
Ceilings ............................................................................... 77
Acoustic leaks ................................................................... 78
Discontinuous construction ............................................ 78
3.9. VIBRATION CONTROL ................................................................... 80
3.10. VENTILATION AND AIR CONDITIONING SYSTEMS .................. 82

CHAPTER 4. CRITERIA FOR NOISE CONTROL A N D

SO UND INSULATION ................... ................................................................... 85

4.1. INTRODUCTION ............................................................................... 85

4.2 HEARING DAMAGE ........................................................................ 86
4.3. NOISE RATING AND NOISE CRITERIA CURVES ....................... 87
4.4. NOISE IN THE HOME ..................................................................... 90
4.5. SOUND INSULATION BETWEEN DWELLINGS ........................... 91
Examples of national recommendations ....................... 92
Britain ................................................................................. 92
Germany ............................................................................. 94
Denmark .............................................................................. 95
Other nations .................................................................... 95

CHAPTER 5. M EASURING TECHNIQUES ................................................. 99

5.1. INTRODUCTION .............................................................................. 99

5.2. GENERALISED CHAIN OF MEASUREMENT .............................. 99
5.3. SOURCES OF SOUND AND VIBRATION ................................. 100
5.4. MICROPHONES AND ACCELEROMETERS ............................. 104
Selection of a microphone ........................................... 104
Preamplifier selection ................................................... 107
Calibration of microphones and associated
measuring systems ..................................................... 109
Selection of an accelerometer .................................... 110
Preamplifier section ....................................................... 112
Calibration of a vibration measuring system ............ 114
5.5. AMPLIFIERS, FILTERS, ANALYSERS AND RECORDERS .... 114
Portable instrum ents for sound measurements ...... 114
Portable instrum ents for vibration measurement .... 116
Measuring am plifiers and filte rs ................................. 118
Frequency analysers ..................................................... 118
Tape Recorders .............................................................. 119
Chart recorders .............................................................. 1 20
Calculators ...................................................................... 121
CHAPTER 6. SUGGESTED INSTRUM ENTATION ....................................122

6.1. INTRODUCTION ............................................................................. 122

6.2. REVERBERATION TIME ................................................................ 122
Portable arrangement - pistol shot method ............... 1 24
Portable arrangement - filtered noise method ........... 125
Automatic arrangement paper loop method .............. 1 26
Autom atic arrangement - digital frequency
analyser / calculator method ..................................... 128
6.3. SOUND DISTRIBUTION ................................................................ 129
Design stage - model techniques ................................ 1 30
Existing room - measurement technique ................... 131
6.4. SOUND ABSORPTION ................................................................. 132
Reverberation room method ......................................... 132
Standing wave method .................................................. 1 34
Tone burst method ......................................................... 138
6.5. SOUND INSULATION .................................................................... 140
Airborne sound insulation ............................................. 140
Field measurements .......................................... 140
Laboratory measurements ................................ 1 43
Impact sound insulation ................................................ 146
Field measurements .......................................... 146
Laboratory measurement ................................. 147
6.5. SOUND POWER ............................................................................ 149
6 6 ROOM MODES .............................................................................. 1 50
6.7. DIRECTIVITY OF NOISE SOURCES ............................................ 152
6.8. VENTILATION AND SERVICE SYSTEM NOISE ........................ 1 52
6.9. VIBRATION MEASUREMENT ..................................................... 153
Locating and m onitoring vibration .............................. 153
Flanking transm ission .................................................... 157
Loss factor ....................................................................... 157
Transmission of shock ................................................... 158
Mechanical m obility ...................................................... 159

APPENDIX ............................................................................................................ 160

Derivation of the normal mode equation for a rectangular room .... 1 60

BIBLIOGRAPHY .................................................................................................. 163

IN D EX ................................................................................................................... 165
 FOREWORD

This booklet is intended as an introduction to the methods and instrum enta­

tion developed by Briiel & Kjær in the sphere of architectural acoustics. W ith
this aim in mind the reader is presented with the basic theory necessary to
make the best use of the instrum ents' capabilities.

As this booklet is an introduction to the subject the amount of mathematics

employed has been kept to a m inim um . The reader w ho requires a more th o r­
ough rendering of a particular topic should consult the reference literature.
The suggested instrum ent arrangements should be considered as a guide
only, for many alternative arrangements are usually possible.

7
 INTRODUCTION

Architectural acoustics can be defined as the study of the generation, prop­

agation and transmission of sound in rooms, dw ellings and other buildings.

Although a relatively new science, architectural acoustics permeates every

w alk of modern life. Correct application of the principles of architectural
acoustics can considerably improve the quality o f life at work, during leisure
tim e and in the home. Some sounds are desirable and need to be enhanced
or emphasized (e.g. music in a concert hall; the speakers voice in a debating
chamber etc), other sounds are highly undesirable (known as noise) and need
to be reduced or prevented (e.g. noise in a factory workshop; noise from a
neighbour's party in the early hours of the m orning etc). In m any countries
m inimum lim its have been set for the permitted noise levels in a particular
environment (e.g. in the home, at the place of work). Regulations have also
been drawn up defining the m inim um acceptable acoustic properties of build­
ing elements (e.g. walls, floors, doors) and the m inim um acceptable sound in­
sulation that should exist between adjoining dwellings.

These regulations are sometimes recommendations and sometimes enforce­

able by law.

Careful thought about the acoustic properties of a proposed building at the

design stage, perhaps in conjunction with the results from acoustic measure­
ments on material samples and scale models, can often save much time and
effort later on. It is frequently the case, however, that alterations have to be
made to improve the acoustic properties of the finished building. To do this ef­
fectively, measurements usually have to be performed before a remedy can
be proposed.

The first chapter of this booklet gives a brief summary of the more import­
ant concepts and definitions necessary for an understanding of architectural
acoustics. In chapter two, the acoustics of rooms are examined in detail.
Chapter three deals w ith the acoustics of buildings and the transm ission of
airborne and impact sound. Chapter four describes some of the more import­
ant noise criteria and regulations which are pertinent to architectural acous­
tics. Chapter five introduces the measuring techniques and chapter six, the in­
strumentation most commonly used in architectural acoustics.

8
 1. FUNDAMENTALS AND DEFINITIONS

1.1. NATURE OF S O U N D

Sound is the sensation perceived by the human ear resulting from rapid
fluctuations in air pressure. These fluctuations are usually created by some vi­
brating object w hich sets up longitudinal wave m otion in the air.

Fig. 1.1. Propagation o f ripples on the surface of a pond

M ost people have some intuitive idea of w hat constitutes a wave. Almost
everyone has seen ocean waves breaking on the seashore or has noticed the
ripples which radiate away from th e place where a pebble strikes the surface

9
of a pond (Fig. 1.1). Sound waves are a particular type of a general class of
waves known as elastic waves. Elastic waves can occur in media w hich pos­
sess the properties of mass and elasticity. If a particle of such a medium is
displaced then the elastic forces present w ill tend to pull the particle back to
its original position. The term particle of the medium denotes a volum e ele­
m ent large enough to contain m illions of molecules so that it may be consid­
ered as a continuous fluid, yet sm all enough so th a t such acoustic variables
as pressure, density and velocity may be thought of as constant throughout
the volume element.

The displaced particle possesses inertia and can therefore transfer momen­
tum to a neighbouring particle. The initial disturbance can therefore be propa­
gated throughout the entire medium.

There are several analogies th a t can be drawn between the propagation of

a sound wave and the propagation of the ripples on the surface of the pond.
Both disturbances travel away from their respective sources at constant
speed. Both disturbances propagate by an exchange of momentum and there
is no net transfer of matter away from the sound source just as there is no
net fluid flow in the pond.

The important distinction is, however, that the ripples are propagated by
transversal waves (i.e. the particle velocity is at rig ht angles to the direction
of propagation) whereas sound in air is propagated by longitudinal waves (i.e.
the particle velocity is in the direction of propagation.

1.2. WAVE TERM INOLOGY

There are a number of terms in common use used to describe the nature of
propagation of a sound wave. Some of the more important terms are defined
here.

Plane Wave

W hen corresponding wavefronts of a sound wave propagate parallel to

each other then the sound wave is known as a plane sound wave e.g. the
sound wave produced by a piston oscillating in a long cylinder.

Diverging Wave

A diverging sound wave is one w here the sound energy is spread over a
greater and greater area as the wave propagates away from the sound source
i.e. the the sound intensity dim inishes w ith distance from the source.

10
Spherical Wave

A spherical sound wave is produced by a sound source w hich radiates

sound energy equally in all directions e.g. a monopole source.

Progressive Wave

When there is a transfer of energy in the direction of propagation of the

sound wave the wave is designated as progressive.

Standing Wave

A standing wave is produced by the constructive interference of two or

more sound waves which gives rise to a pattern of pressure maxima and m i­
nima which is stable w ith tim e e.g. standing waves can exist in tubes, m usi­
cal instruments, organ pipes and in larger volumes such as rooms.

1.3. FREQUENCY OF SO U N D

Consider the air near to the surface of some vibrating object e.g. piston in
a tube, tuning fork, loudspeaker diaphragm. The series of compressions and
rarefactions produced by the movement of the object constitute a sound
wave, the frequency of w hich is determined by the rate of oscillation of the
object. When the oscillation repeats itself, th e motion is said to have com­
pleted one cycle. The number of cycles per second is called the frequency, f.
The unit of frequency is the Hertz. 1 Hertz = 1 cycle/sec. The tim e taken for
the oscillation to repeat itself is known as the period, T.

( 1. 1)

Fig. 1.2 shows the relationship between the compressions and rarefactions
and the pressure variation produced by a vibrating piston.

It is often useful to express frequency in term s of the angular frequency.

For a vibration of frequency, f, the corresponding angular frequency, <j , is

œ = 2nf ( 1 .2 )

11
 F/g. 1.2. Compressions and raréfactions in a longitudinal plane wave
a) Pressure variation o f a sinusoidal wave
b) Propagation of a plane wave in a tube

1 .4. SPEED OF S O U N D

The speed of propagation or speed of sound is dependent on the mass and

e la sticity of the medium. The elasticity of air is determined by experiment to
be a constant m ultiplied by the atmospheric pressure. The constant, y , is
found to be the ratio of the specific heat of the air at constant pressure to the
specific heat at constant volume. For the temperature range that acousticians
usually deal with this ratio is 1,4. The speed of sound, c, in air is given by

1,4 P n
c = (1 3 )

w here P0 = atmospheric pressure

p = density of air

Assum ing that air acts as a ideal gas, it can be show n that the speed of
sound depends only on the absolute temperature of the air according to the
equation,

12
 (1.4)
c ~ 332J 1* m

w here t = air tem perature (C°)

c = speed of sound (m.s —1 )

A t ordinary room temperatures, the speed of sound is approximately

3 4 0 m .s- 1 .

1.5. WAVELENGTH

The wavelength, A, is the distance between tw o successive pressure max­

ima or between successive pressure minima in a plane wave. The relation­
ship between X, c and f is

c = y .f (1-5)

The nomogram in Fig. 1.3. gives the wavelength corresponding to a particu­

lar frequency.

1.6. DISPLACEMENT

The distance between the instantaneous position of a vibrating particle and

its mean position is the displacement of the particle.

1.7. AMPLITUDE

The maximum displacement experienced by a vibrating particle is known as

the amplitude of vibration. The amplitudes of vibration of airborne sound
waves which occur in practice are very small. The amplitudes range from

13
about 1 0 - 7 mm up to a few mm. The smaller amplitude corresponds to the
sound which is just perceptible by the ear w hile the greater am plitude is the
lim it beyond w hich the ear w ould suffer damage.

1.8. PARTICLE VELOCITY

The differentiation of the displacement of the particle with respect to time

yields the particle velocity. This property of a sound wave can be measured di­
rectly using a classical technique developed by Lord Rayleigh known as Ray­
leigh's disc. The method is, however, tedious and is of little practical import­
ance.

1.9. SOUND PRESSURE

The pressure variations produced when a sound wave propagates through

the air are very small compared w ith the static atmospheric pressure. The
slightest sound that an average young adult can detect corresponds to a
sound pressure of 0 ,0 0 0 0 2 Pa (N.B. 1Pa = 1N .m ~ 2 ). This sound pressure is
superimposed on the ambient atmospheric pressure which is of the order of
105 Pa. The concept of sound pressure is extremely important because of all
the quantities w hich could be used to characterise the "strength" of a sound
wave (e.g. particle velocity, sound intensity) it is the sound pressure which is
the most amenable to measurement.

1.10. SPECIFIC ACOUSTIC IM PEDANCE

For a sinusoidal wave form, the ratio of the acoustic pressure in a medium
to the associated particle velocity is defined as the specific acoustic imped­
ance of the medium for the particular type of wave motion present. The spe­
cific acoustic impedance, z, for plane waves is a real quantity (i.e. not com­
plex) of magnitude pc where p is the density of the medium and c is the
speed of sound in the medium. The SI unit of specific acoustic impedance is
the rayl, expressed in N.s.m- 3 . As the product pc is a characteristic property
of the medium, this product is also referred to as the characteristic imped­
ance of the medium. At normal temperature and pressure (i.e. 20°C and
1 0 5 Pa), the specific acoustic impedance of air has the value of 41 5 rayls.

Although the specific acoustic impedance of a medium is a real quantity for

progressive plane waves, this is not true for standing plane waves or for di­
verging waves. In general, z, w ill possess both a real part, r, and an imagi­
nary part, jx, so that

z = ? = r + jx (1.6)

14
where r = specific acoustic resistance
x = specific acoustic reactance of the medium for the particular
wave motion being considered.

The concept of acoustic impedance is im portant when dealing w ith stand­

ing waves and absorption of sound energy as for example in the application
of the wave equation to certain problems in room acoustics (see Chapter 2).

1.11. DECIBEL

In theorectical investigations of acoustic phenomena, it is convenient to

express sound pressure, sound intensity and sound power by Pa, W .m .—2
and W respectively. For practical measurements, however, it is usual to exp­
ress these quantities using logarithmic scales. There are several reasons for
using such scales. One is a consequence of the very w ide range of sound
pressures and intensities encountered e.g. the range of audible intensities is
from 10—12 to 10W.m—2 . The use of a logarithmic scale compresses the
range of numbers required to describe th is wide range of intensities. Another
reason is that the human ear subjectively judges the relative loudness of tw o
sounds by the ratio of th e ir intensities, w h ich is a logarithmic behaviour.

The most commonly employed logarithm ic scale for describing sound levels
is the decibel scale. One decibel is the energy or power ratio, r, defined by

L o g ]0 r = 0,1 (1.7)

For sound pressure or particle velocity ratios, the definition is Log10r =

0,05.

The important point about the decibel is th a t it is a relative measurement and

that each quantity measured in decibels is expressed as a ratio relative to a
reference pressure, pow er or intensity or whatever other quantity is being
considered. It is good policy to employ the word "le v e l" whenever decibels
are used e.g. sound pressure level, sound power level etc. as a reminder that
the measurement is a ratio relative to some reference level.

1.12. PEAK, AVERAGE A N D RMS

Whenever measuring a quantity w h ich varies w ith tim e, such as sound

pressure level, it is necessary to state w he th er the measurements represent
the peak, average or rm s value of the signal. The relationship between these
three measurements for a sinusoidal signal is shown in Fig. 1.4.

The rms (root mean square) value is the most commonly used because it

15
has a direct relationship to the energy content of the signal. The rms value of
a signal is defined as

A rms= f \ a 2( t) d t
7 ( 1.8)
'V / ■ ()

The average value of the signal is

(1.9)

The peak value, A peak , is the maximum amplitude value that the signal
reaches w ith in the period of tim e T.

For the special case of a pure sinusoid the relationship between these three
values is

A r m s 2 ^ 2 /^ \ a '/ e r a e e \ — 2 ^ peak (1 1 0 )

Fig. 1.4. Relationship between rms, peak and average value o f a sinusoidal
signal

1.13. ENERGY DENSITY

A sound wave contains kinetic energy, as a consequence of the particle ve­

locity, and potential energy as a result of the sound pressure. This energy
propagates w ith the speed of sound. The sound wave, therefore, conveys me-

16
chanical energy. The amount of energy per u nit volume of a sound wave is
measured by a quantity known as the energy density. For a plane sound
wave the energy density, E, per u nit volume is defined by

2
E = P __m ns ( 1 1 1 )
p c2

where P2 rms = mean square sound pressure (Pa)

p = density of air (kg.m —3 )
c = speed of sound (m.s.—1)
E = energy density (W.s.m- 3 )

1.14. INTENSITY

The intensity, I, of a sound wave is defined as the mean value of the acous­
tic energy which crosses a unit area perpendicular to the direction of propaga­
tion in unit tim e. Unlike the expression for the energy density of a sound
wave, the expression for the intensity is different for different types of sound
field.

For any free progressive wave

(1 12 )
pc

The intensity of a diffuse sound field at the w alls of a room is

4pc

where p2 rms = mean square sound pressure (Pa)

p = density of air (kg.m —3)
c = speed of sound (m.s.- 1)
I = intensity (W .m.- 2 )

1.15. INTENSITY LEVEL

The intensity level, IL, of a sound of intensity, I is defined by

IL = T O /o g ,0( £ ) (1.14)

where IL is expressed in decibels and l0 is the reference intensity usually

taken as 1 0 ~ 12 W .m .—2 .

17
1 .1 6 . SOUND PRESSURE LEVEL

The sound pressure level, SPL, of a sound of root mean square pressure
Prms is defined by

SPL = 2 0 l o g , 0( ^ \ (1.15)
' Po '

w here SPL is expressed in decibels and P0 is the reference sound pressure le­
vel of 0,00002Pa. This value is chosen as it approximately corresponds to
the sound pressure of the slightest sound an ear can detect in quiet surround­
ings. Sound pressure level is the quantity that is actually measured w hen a
microphone is placed in a sound field. The portable instrum ent used to mea­
sure SPL is known as a sound level meter. Some typical sound pressure le­
vels are shown in Fig. 1.5.

Sound Pressure ( Pa) Sound Pressure Level (dB re 2 x 10’ ^Pa)

-140
100 000000 -
-130
Near to a jet aircraft ta k in g o ff
-120
10000000 -
-110
-100 Near a pneumatic d r ill
1000000-
•90
Inside a m otorcar
■80
100000-
■70

60 General office
10000 -
50

40 Q uiet livingroom
1000-
30

20
100- Q uiet countryside
10

20- 0 Threshold o f hearing

Fig. 1.5. Some typical sound pressure levels

18
1 .17 . SOUND POWER LEVEL

The sound power level, SWL or Lw is a measure of the energy output of a

sound source. The SWL is defined by

SWL = 10 lo g , (1.16)

where SWL is expressed in decibels, W is the acoustic power of the source

and W 0 is the reference acoustic power of 1 0 —12 W. When assessing the
noisiness of a machine or domestic appliance, it is not sufficient to quote just
the sound pressure level measured using the A-w eighting network since the
level measured is dependent on the environment. This dependence has been
known for many years and consequently it has generally been accepted that
the sound power emitted by the machine should be determined as this is the
fundamental indication of the noise output and is virtually independent of the
environment. Some typical sound power outputs of various sources are
shown in Fig. 1 .6.

Power (Watts) Power Level

(dB re 10“ 12 W)

100,000,000 2 0 0 -j Saturn ro c k e t (5 0 ,0 0 0 ,0 0 0 W)

1,000,000 180-

10,000 160- 4 Jet A irlin e r (5 0 ,0 0 0 W)

100 140-
Large orchestra (1 0 W)

1 120- C h ipping ham m er (1 W)

0,01 100-

Shouted speech (0,001 W)

0,000,1 80-
C onversational speech (20 x 1 0 " 6 W)
0,000,001 60 -

0,000,000,01 40 -
W hisper ( 1 0 -® W)
0,000,000,000,1 20-

0,000,000,000,001 0 -
771111

Fig. 1.6. Some typical sound pow er levels

19
1 .1 8 . SO UND SOURCES - THEORETICAL AND PRACTICAL

The variety of sound sources one has to deal w ith in architectural acoustics
is very large e.g. loudspeakers, machines, the voice, musical instrum ents.
The characteristics and the directivity pattern of the sound generated by each
type of source may vary considerably. To describe each type of sound source
comprehensively in theoretical term s would be very tedious and involved. It is
fortunate, therefore, th a t many sound sources can, to a good degree of accu­
racy be approximated by one or a combination of several idealised theoretical
sources such as the monopole or the dipole (Ref. 1).

1 .19 . MONOPOLE OR SIMPLE S O U N D SOURCE

The simplest type of sound source for generating spherical acoustical

waves is a pulsating sphere. Due to it's symmetry, th is simple or monopole
source w ill produce harmonic spherical waves in any surrounding medium
that is homogeneous and isotropic. M any practical sources behave, at least
to a firs t approximation, like a monopole providing th e ir dimensions are small
compared with the wavelength of the radiated sound (Fig. 1.7).

As the distance from the source, increases, the area over which the em it­
ted energy is spread is increased and consequently the resulting sound inten­
sity is reduced.

For a monopole radiating harmonic spherical waves the sound intensity is

20
 (1.17)

where I = Sound intensity (W.rn- 2 )

W m = Sound power (W)
r = Distance from monopole (m)

Taking logarithms to the base 10 and using the relationship between sound
intensity and sound pressure yields

SPL = SWL - 2 0 lo g , 0 (a) - W lo g , 0 (4n)

(1 1 8 )
SPL = SWL - 20 lo g , 0 ( 0 - 11dB

where SPL = Sound pressure level re 2 * 1 0 —5 Pa

SWL = Sound power level re 1 0—12 W att
r = Distance from the source in metres

Therefore, for spherical radiation the SPL decreases by 20log (2) = 6dB
each time the distance from the source is doubled. This result is known as
the inverse square law.

The mean square pressure measured at a distance r from a sphere of mean

radius a and pulsating w ith a frequency f, assuming that ka « 1 rad, is
given by
2
pckQ
16 nr (1.19)

where k = wave number (rad m—1)

c
c = speed of sound (m .s ~ 1)
p = density of air (Kg.m—3)
Q = strength of the source (m3 .s—1)
r = distance from the source (m)

The strength of the source, Q, is defined as the product of the surface area
and the rms velocity amplitude of the surface, U, that is

Q = 4 na 2U (1 2 0 )

N.B. The quantities Q, p, U must all be of the same type i.e. all peak or all
rms.

The intensity, I, of the sound waves produced by a monopole source is

21
 I = p h ™ ( 1 2 1 )

pc

From the above equations it is seen that both the mean square pressure and
the intensity decrease as one moves away from the source by an amount pro­
portional to the square of the distance from the source.

The sound power, W m , radiated by the source is

Wm = 0 . 22 )

which is independent of the radial distance as required by the principle of the

conservation of energy.

Probably the simplest practical source w hich behaves like a monopole is a

loudspeaker mounted in a suitably absorbant enclosure and situated in free
space. W hen the wavelength of the sound produced is much greater than the
dimensions of the source i.e. when k a « 1 , the sound pressures measured at
distances from the source such that r » a , are found to be the same for all
sources of equal strength regardless of the shape of the radiating surface.

1.20. DIPOLE

Consider a loudspeaker mounted in an open backed cabinet. The radiation

pattern produced w ill be the result of the radiation from both sides of the
loudspeaker cone. As the dimensions of th e baffle become small compared to
the wavelength of the sound emitted then the radiation pattern approaches
that of a dipole or acoustic doublet. The situation can be thought of as two
monopoles 1 80° out of phase, of source strength Q separated by a small dis­
tance I ( k .l« 1 ) . The sound pressure, pd , produced by such a system at a
point p (r,0 ) is

p = P ^ !f9 J . cos g (1.23)

a 4 nr

The quantities Pd and Q m ust be of the same type i.e. all peak or all r.m.s.
This form ula is only valid fo r k r » 1 .

The sound pressure pattern radiated by a dipole is shown in Fig.1.8. The

sound pressure field is symmetrical about the axis joining the two monopoles
which in practice is the axis of the loudspeaker.

22
 ^ ______ ^ ___ p(r,0)

A xis o f
dipole

-Q # | I # + Q I
I H*----- JV ---- ►»

780505

Fig. 1.9. Variation of sound pressure level w ith distance from a dipole

The variation of sound pressure as a function of distance from the dipole is

shown in Fig. 1 .9. Close to th e dipole (the near field) the sound pressure de­
cays rapidly, pcc1/r2, while fa r from the dipole (the far field) the sound pres­
sure decays more slowly, poc1/r. Providing th a t the sound pressure is mea­
sured in the far field, the intensity of the dipole source, ld, may be w ritten as

23
The sound power output of the dipole, W d, is given by

(1.25)

w here Q is an r.m.s. value.

In the derivation of the foregoing equations it is assumed that the distance

I is very small i.e. I « 1 / k .

On comparing the sound power of the monopole and the dipole using equa­
tions 1.22 and 1.25, one finds that

Wd Qd2
( 1.26)
Wm 3 Qm*

W hen the source strengths are equal (i.e. Qm = Q di) then since the wave
number, k, is proportional to the frequency, it can be seen that at low fre ­
quencies the dipole is far less efficient in radiating sound energy than the
monopole. This is the reason why a loudspeaker m ounted in a cabinet can
reproduce low frequencies whereas a loudspeaker w ith o u t a baffle cannot.

Practical sources w hich radiate like a dipole include a loudspeaker w ith o u t

a baffle, pure tone fans, a tuning fork and a vibrating beam.

Monopoles and dipoles have been employed to form more complex th e o re ti­
cal sources such as the lateral quadrupole and the longitudinal quadrupole
but as these have no direct application in architectural acoustics they w ill not
be discussed here.

1 .2 1 . DIRECTIVITY PATTERN

M any sound sources that are met in practice are more complicated than
either the monopole or the dipole. Their behaviour has to be determined by
measurement rather than by theoretical prediction. Measuring the sound
pressure level at a fixed distance from a source, but in different directions,
w ill generally yield d ifferent levels. The plot of these levels on a polar d ia ­
gram is known as the directivity pattern of the source. A surface can th e re ­
fore be described around a source over w hich the same sound pressure level
exists. However, it is usually sufficient to specify the directivity pattern in the
vertical and the horizontal directions Figs. 1 .10 and 1.1 1).

D irectivity patterns are extremely useful for supplying a lot of inform ation
about the nature of a sound source in a form which is easy to assimulate.

24
Fig. 1.10. Typical horizontal directivity patterns for the Sound Power Source
Type 4 2 0 5

Fig. 1.11. Typical vertical directivity patterns for the Sound Power Source
Type 4 2 0 5

These patterns can and usually do vary w ith frequency; most sources being
highly directive at high frequencies and non-directive or nearly so at low fre ­
quencies.

Besides describing the nature of such sources as loudspeakers, reference

sound sources, machines etc., the concept of directivity patterns is used for
describing the pressure response of microphones and sound level meters.

25
1 .2 2 . DIRECTIVITY FACTOR AND DIRECTIVITY IN D EX

The directivity of a source can be specified either by the directivity factor,

D, w hich is a dimensionless quantity defined by

0 = (1.2 7)
'r e f

or by the directivity index, d, expressed in dB and defined by

d = 1 0 /o g ,0 D = 10 lo g T o ijt - j (1.2 8)

In these equations. I, is the intensity measured at some distance from the

source in the direction in w hich the directivity is to be specified and lref is a
reference intensity defined by

W
= ^ (1-29)

w here W = sound power output of the source.

Values of the directivity factor range from unity for the case of the monopole
to large numbers for highly directive sources. The directivity also depends on
the source position. Table 1.1 gives the directivity factor and directivity index
for a monopole in various locations.

Source Location D irectivity Directivity

Factor Index, dB

Free field 1 0
e.g. suspended between floor
and ceiling

On a fla t plane 2 3
e.g. on the floor

A t junction of tw o perpen. 4 6
planes e.g. floor and wall

A t junction of three perpen. 8 9

planes e.g. in a corner
780118

Table 1.1. Directivity factor and directivity index for a monopole in various lo ­
cations

26
 From a knowledge of the sound power and the directivity pattern of a
source, one can estimate the sound pressure levels generated by the source
in almost any given acoustic environm ent. Such estimations can be used to
position noisy machines in workshops and offices etc., where they would
cause the least disturbance.

1 .2 3 . SO UND FIELD OF A SO UND SOURCE

The nature of the sound fieia around a sound source in a room consists of
tw o components i.e. the direct field and the reverberant field. (Fig. 1.1 2). The
immediate vicinity of the source is know n as the near field. In this region the
particle velocity is not necessarily in the direction of propagation of the
sound wave. Furthermore, the sound pressure may vary considerably w ith po­
sition and the sound intensity is not sim ply related to the mean square pres­
sure. The extent of the near field is d ifficu lt to define as it depends on many
factors such as the frequency, dim ensions of the source and phases of the ra­
diating surfaces. (Fig. 1.13).

Distance fro m Source, m 771200

Fig. 1.12. Combination o f direct and reverberant fields

In the region known as the far fie ld the sound pressure level decreases
6 dB each tim e the distance between the measuring microphone and the
source is doubled providing the source is in free space which can be simu-

27
Fig. 1.13. Description o f the sound field around a sound source in a reverber­
ant room

lated in an anechoic room. In this region, the particle velocity is principally in

the direction of propagation and the intensity is proportional to the mean
square pressure.

If the source is radiating into a normal room then reflections of the sound
waves from the room's boundaries create a reverberant field w hich is super­
imposed on the source's far field. In a highly reverberant or "live" room, the
reverberant field may swamp the far field altogether. The reverberant field is
referred to as a diffuse field if the sound energy density in this region is very
nearly uniform.

For a spherically radiating source it has been shown that

where l^ = sound intensity due to the direct field

The sound intensity in the reverberant field, l r, can be expressed as

, 4W
'r = - n (1-31)

28
where R is the room constant defined by

(1.32)

where S = total surface area of room

â = average room absorption coefficient

Combining these two expressions yields

WD_ 4W
(1.3 3)
~ 4nr2 + ' R

where D = directivity factor

Taking logarithms to the base 10 of this equation and expressing the result
in terms of sound power gives

(1.34)

where SPL = sound pressure level re 2 x 1 0 ~ 5 Pa

SWL = sound power level re 10—12 Watt

Obviously, the closer to the source, the greater the effect of the direct sound
on the measured sound pressure level. A t a certain distance from the source,
however, the contributions to the sound pressure level from both the direct
and reverberant field w ill be equal. This distance, known as the room radius,
can be calculated by setting

therefore the room radius, r r, is given by

1.36)

1.24. THE EAR

Some knowledge of the ear's charactertistics is necessary to fully appre­

ciate the problems w hich arise in architectural acoustics. The range of sound
pressures to which the ear can respond is prodigious. The ear can w ithstand

29
sounds of pressure amplitudes in excess of 100 Pa and yet can detect sound
pressures of 0 ,00001 Pa. Such small sound pressures, in the ear's most sen­
sitive range w hich is from 1 00 0 Hz to 5 0 0 0 Hz, produce a displacement of
the eardrum of the order of 1 0 —11 m. This m inute distance is approximately
one tenth of the diameter of a hydrogen molecule. The ear, however, is more
than just an extremely sensitive microphone. It is also, together with the
brain a frequency analyser capable of fine discrim ination between tones.

The m inim um intensity level perceptable by the ear at a particular fre ­

quency is known as the threshold of hearing or threshold of audibility at that
frequency. The threshold of hearing varies from person to person even
among people who have "n o rm a l" hearing. This threshold is also a function
of the age of the listener, the progressive loss in sensitivity at the higher fre ­
quencies w ith age is called presbycusis. From Fig. 1.14 one can see that it re­
quires nearly a m illion tim es more power to produce an audible sound at
50 Hz than at 3 0 0 0 Hz.

Fig. 1.14. Audible range o f frequencies and sound pressure levels bounded by
threshold o f hearing and the threshold of feeling together with ap­
proximate regions for speech and m usic

30
 As the intensity of the acoustic waves incident on the ear is increased, the
sound perceived by the ear becomes louder and louder u ntil the sensation
ceases to be one of hearing but one of "tic k lin g " or feeling w ithin the ear.
This level is known as the threshold of feeling. It is less dependent on fre ­
quency than the threshold of hearing and has a value of about 120dB. The
approximate limits of the frequencies and the intensities normally experi­
enced in speech and music are also shown in Fig. 1.14.

1.25. LOUDNESS, PHONS A N D SONES

The subjective characteristics of a sound known as the loudness is a fu n c ­

tion of the sound's intensity and frequency. From Fig. 1.1 4 fo r example, one
can see that a pure tone having an intensity level of 20 dB and a frequency
of 1000 Hz would be clearly audible whereas a tone having the same inten­
sity but a frequency of 1 00 Hz would not be heard at all as it lies well below
the threshold of hearing.

From the results of many subjective experiments, contours of equal loud­

ness can be drawn on an intensity vs frequency diagram as shown in
Fig. 1.15.

The unit of loudness level is the phon. The loudness level in phons of any
sound is defined as being num erically equal to the intensity level in decibels
of a 1 0 0 0 Hz tone that is judged by the average observer to be equally loud.

As the apparent loudness of a sound is not directly proportional to the

sound's loudness level, subjective experiments have been performed in order
to establish a scale on w hich a doubling of the number of loudness units dou­
bles the subjective loudness and trebling of the number of loudness units tre ­
bles the subjective loudness and so on.

The unit o f loudness is the sone. A sone is defined as the loudness of a

1000 Hz tone having an intensity level of 4 0 dB, i.e. a sone is equal to the
loudness of any sound having a loudness level of 40 phons.

No instrum ent exists w hich can actually measure the loudness of a compo­
site sound w hich contains many frequency components. However, sound le­
vel meters can make precise measurements of the sound pressure level of
such sources from which reasonably accurate estimates of the loudness can
be made.

31
 FREQ UENCY Hz

7804/8

Fig. 1.15. Equal loudness contours

32
 1.26. S O U N D LEVEL METERS AND W EIG H TIN G NETWORKS

A sound level meter consists essentially of a high quality microphone, a lin ­

ear am plifier, one or more attenuators, a set of frequency weighting ne­
tworks including a linear response and an indicating m eter or an LED (light
emitting diode) digital display.

The purpose of the w eighting networks is to make the readings on the in d i­

cating meter correspond as closely as possible to the perceived loudness le­
vels. The frequency characteristics of these weighting networks are therefore
approximately the inverse of the corresponding equal loudness contours
shown in Fig. 1.16. Ideally the sound level meter would contain a netw ork
possessing a frequency response which w ould be continuously variable over
the dynamic range. In practice, however, it is sufficient to use only a fe w
fixed networks. These w ere originally defined as:

A-w eighting: used for loudness levels below 55 phons.

B-weighting: used for loudness levels between 55 and 85 phons.
C-weighting: used for loudness levels over 85 phons.
D-weighting: used to account for the increase in annoyance produced
by the high frequency w hine present in the noise pro­
duced by certain aircraft.

However, more recent work has not substantiated these historical associa­
tions so that the frequency weightings are now largely conventional. F urther­
more, the A-weighting is now frequently specified for rating sounds irrespec­
tive of level and is no longer restricted to low level sounds.

Fig. 1.16. Frequency response o f the weighting networks

33
 2. ACOUSTICS OF ROOMS

2.1. D EFIN ITIO N OF ROOM ACOUSTICS

Consider a sound source w hich is situated in a room. Sound waves w ill

propagate away from the source until they encounter one of the room's
boundaries w here, in general, some of the sound energy w ill be reflected
back into the room, some w ill be absorbed and some w ill be transmitted
through the boundary. The complex sound fie ld produced by the multitude of
reflections and the behaviour of this sound field as the sound energy in the
room is allowed to build up and decay constitutes the acoustics of the room.

2.2. GEOMETRICAL ROOM ACOUSTICS

If one can assume that the dimensions of a room are large compared to the
wavelength of sound then one may treat the sound waves in the room in
much the same way as light rays are treated in geometrical optics. This situa­
tion frequently occurs in architectural acoustics. In analogy w ith light rays,
sound rays are reflected from hard plane w a lls in accordance w ith the laws
of reflection i.e. the incident ray, the reflected ray and the normal to the sur­
face at the point of incidence all lie in the same plane; the angle of incidence
is equal to the angle of reflection (Fig. 2.1). Therefore sound rays incident on

34
a curved surface w ill either be focused or dispersed depending on w hether
the surface is concave or convex (Fig.2.2). Diffraction of sound rays can and
does occur but the effect is more noticeable for low frequency, long w ave­
length sounds than w ith high frequency sounds of short wavelength.

Fig. 2.2. Reflections o f sound rays

Fig. 2.3. Graphical construction of the firs t reflections of the sound waves in
a concert hall

The concept of a sound ray and the geometrical study of sound ray paths
plays an important role in the design of large rooms and auditoria, enabling
troublesome echoes and flu tte r effects to be detected and dealt with at the

35
stage of designing the building. Fig.2.3 shows how geometrical constructions
can be used to position sound reflectors on the ceiling of a concert hall in or­
der to improve the distribution of sound. A lim itation of the geometrical ap­
proach is that usually only the primary and possibly the secondary reflections
can be studied before the sound ray being follow ed becomes "lo s t" in the re-
verberent sound field.

2.3. GROW TH AND DECAY OF SO UND IN A ROOM

When a sound source is placed in a room, the sound intensity as measured

at a particular point w ill increase in a series of small increments, due to the
reflections arriving from the walls, floor and ceiling, until an equilibrium posi­
tion is attained where the energy absorbed by the room is equal to energy ra­
diated by the sound source. When the sound source is abruptly switched off
the sound intensity in the room w ill not suddenly disappear but w ill fade
away gradually, the rate of decay being prescribed by the amount and posi­
tion of the absorbing material in the room. This lingering of the sound is
known as reverberation. The rate of absorption of sound energy in the room
w ill be, in the main, proportional to the sound intensity so that the growth
and decay of sound pressure in the room is an exponential function of tim e
(Fig.2.4.).

If one measures the sound pressure levels in dB in a decaying reverberant

field as a function of tim e then one obtains a reverberation curve which is us­
ually a fairly straight line although the exact form depends on many factors in ­
cluding the frequency spectrum of the sound source and the shape of the
room.

36
2.4. REVERBERATION TIM E

At the beginning of this century W.C. Sabine carried out a considerable

amount of research on the acoustics of auditoria and arrived at an empirical
relationship between the volume of the auditorium , the am ount of absorptive
material w ith in the auditorium and a quantity which he called the reverbera­
tion tim e. This relationship is now known as the Sabine form ula:

0,161 y
RT = ( 2 . 1)
A

where RT = the reverberation time defined as the tim e taken for a

sound to decay by 60 dB after the sound source is
abruptly switched off
V = the volume of the auditorium in m3
A = the total absorption of the auditorium in m2 -sabins.

The absorption unit of 1m2 -sabin represents a surface capable of absorbing

sound at the same rate as 1m2 of a perfectly absorbing surface e.g. an open
window.

2.5. ABSORPTION COEFFICIENT

Material Frequency, Hz
125 250 500 1000 2000 4000

Air, per cu. m. nil nil nil 0,003 0,007 0,02

Acoustic paneling 0,15 0,3 0,75 0,85 0,75 0,4
Plaster 0,03 0,03 0,02 0,03 0,04 0,05
Floor, concrete 0,02 0,02 0,02 0,04 0,05 0,05
Floor, wood 0,15 0,2 0,1 0,1 0,1 0,1
Floor, carpeted 0,1 0,15 0,25 0,3 0,3 0,3
Brickwall 0,05 0,04 0,02 0,04 0,05 0,05
Curtains 0,05 0,12 0,15 0,27 0,37 0,50

Total absorption of
one seated person 0,18 0,4 0,46 0,46 0,51 0,46
780120

Table 2.1. Typical absorption coefficients of 1m 2 o f materia!

37
 The absorption coefficient of a material, as originally defined by Sabine, is
the ratio of the sound absorbed by the m aterial to that absorbed by an equiva­
lent area of open window hence the absorption coefficient of a perfectly ab­
sorbing surface would be 1 .Providing one knows the superficial areas and
the absorption coefficients of the various materials to be used, the reverbera­
tion tim e of an auditorium can be determined at the design stage. To fa c ili­
tate such calculations sets of tables have been published giving the absorp­
tion coefficients of the commonly used building materials as a function of fre ­
quency. The variation in the reverberation tim e of specially designed reverber­
ant rooms (also known as live or hard rooms) as one introduces or removes
absorptive material is a standard method for determining absorption coeffi­
cients (refer to the International Standard ISO 354). (See Chapter 6).

2.6. DERIVATIO N OF FORMULAE FOR REVERBERATION TIM E

Theoretical derivations of Sabine's form ula are usually based on geometri­

cal acoustics utilising the assumptions that the sound in the enclosure is d if­
fuse and that all directions of propagation are equally probable. This is a
gross sim plification of the actual behaviour of sound in enclosures because it
neglects such important factors as room modes, the positioning of absorptive
material, the influence of the shape of the room and others. For fairly reverb­
erant rooms w ith a uniform distribution of absorptive m aterial, Sabine's fo r­
mula gives a good indication of the expected behaviour of sound in the room.
As a room becomes more and more "dead" i.e. the boundaries become more
and more absorbent, so the results obtained from employing Sabine's fo r­
mula become more and more inaccurate. In the limiting case of a completely
dead or anechoic room w here the absorption coefficient of the boundaries is
unity then the reverberation tim e is obviously zero because a reverberant
field cannot exist in these conditions. However, for this situation Sabln's fo r­
mula gives a fin ite value of 0,161 V /A for the reverberation tim e. Several d if­
ferent approaches have been used to derive equations w hich give values of
reverberation time which are in better agreement w ith the measurement re­
sults for rooms containing little absorption. As examples, tw o such formulae
are quoted here.

Eyring's formula for reverberation time is

0,161 1/
( 2 .2 )
RT — S ln(1 — a )

where

_ _ « t S -i + «2^2 + • • • + y -n S n
the mean absorption coefficient of the room.
S i + s2 + ■ • • + sn

38
 + sn the areas of the various materials

a-,, cx2 , • • • the respective absorption coefficients.

Eyring's formula gives results w hich are in much better agreement w ith the
measured reverberation times for dead rooms than Sabine's form ula. Also,
the Eyring formula gives the correct value of RT = O for an anechoic room
i.e. fo r a = 1. One drawback of this improved form ula is that it is only strictly
valid for rooms w hich have the same value of a for all boundaries.

The theory of M illington and Sette leads to the form ula

RT = £ - 0s,' - ln (T -------
r -(1
1 x,)Y (23)

w here Si = the area of the ith material

«j = the absorption coefficient of the ith m aterial.

W hen the materials in a room have a wide variety of absorption coefficients

then the best predictions of reverberation times are obtained by employing
the M illington and Sette formula. This formula can be obtained by s u b stitu t­
ing the effective sound absorption coefficient ae = — ln(1—«¡1 into Sabine's
form ula.

M illington and Sette's formula indicates that highly absorbing m aterials are
more effective than would be anticipated in influencing the reverberation
tim e. Forexample, when the absorption coefficient of amaterial is greater
than 0 ,6 3 then the effective absorption coefficient isseen to begreater than
one.

2.7 . WAVE THEORY OF ROOM ACOUSTICS

Although very useful in certain applications, the geometrical approach to

room acoustics is not a satisfactory method for attempting to explain the be­
haviour of sound w ith in an enclosure. A more adequate but more d iffic u lt ap­
proach is based upon wave acoustics, that is, upon the motion of sound
waves w ithin a three dimensional enclosure. This method is characterised by
the establishing of boundary conditions which describe mathematically the
acoustical properties of the walls, ceiling and the other surfaces in the room.
The difficulty involved in determining these boundary conditions for rooms of
irregular shape e.g. churches, auditoria, rooms containing fu rn iture etc.,
means that an analysis employing the wave theory can only be performed ex­
actly for a number of idealised situations. Although the practical application
of the wave theory is limited, an understanding of the theory is essential in
order to appreciate many of the problems which arise in the acoustics of

39
rooms, e.g. w hy the frequency response of a loudspeaker varies from room to
room; w hy there are pressure maxima and m inim a w ithin a room where
there is a steady sound source.

When employing the wave theory, a room is considered as a complex reso­

nator possessing many normal modes of vibration and each mode having a
characteristic frequency of damped free vibration. These modes can be ex­
cited by introducing a sound source into the room. The acoustic energy sup­
plied by the source can be considered as residing in the standing waves esta­
blished in the room. The characteristic frequencies of vibration of the stand­
ing waves depend principally on the room's size and shape whereas the
damping of these waves depends mainly on the boundary conditions. Using
the simplest possible boundary conditions of rigid w alls and no damping, one
can derive the characteristic frequencies of a room. The effect of damping
can then be taken into account by considering it to be a modification of these
simple conditions.

For a rectangular room a simple relationship exists between the room di­
mensions, lx, ly and lz, and the frequencies corresponding to the normal
modes of vibration of the room (See Appendix for the derivation). This rela­
tionship is

A normal mode of vibration occurs for every permutation of n x, nY and nz

w here less indices have the integer values 0, 1, 2, 3 ...etc. An example of the
use of this expression is given in Table 2.2 w here the characteristic frequen­
cies below 100 Hz have been calculated from eqn.2.4 taking c as 340
m .s.- 1 , for the large reverberation room at the Danish Technical High-
school. The dim ensions of the room are 7,85 x 6 ,2 5 x 4,95 m. The room, act­
ing as a resonator, responds strongly at these resonant frequencies. The nor­
mal modes of a room are not distributed evenly over the frequency spectrum.
A t low frequencies there are very few modes but the number of modes in­
creases rapidly w ith frequency. The modes can occur in bunches around a
particular frequency e.g. in Table 2.2 the characteristic frequencies 2,0,0
and the 0,1,1 modes are very close to each other w hile there is a consider­
able gap between the 43,5 Hz and the next characteristic frequency w hich oc­
curs at 4 8 ,9 Hz. It is this "p refe re n tial" treatm ent of certain frequencies by
rooms w hich affects the output of loudspeakers. Every room imposes its own
characteristics on those of any sound source present, so that the fluctuations
in sound pressure which occur as a microphone is moved from one position
to another, or as the source frequency is varied, may completely obscure the
true output characteristics of the source. If the response curves of loudspeak-

40
 n n n f (Hz)
X y z

1 1 2 77,0
2 2 1 77,6
3 1 1 78,4
2 0 2 81,2
0 3 0 81,6
1 3 0 84,4
3 2 0 84,7
2 1 2 85,6
4 0 0 86,6
0 2 3 87,6
0 3 1 88,5
1 2 2 90,3
4 1 0 90,8
1 3 1 91,1
3 2 1 91,4
2 3 0 92,4
4 0 1 93,2
3 0 2 94,5
4 1 1 97,1
2 2 2 97,7
3 1 2 98,4
2 3 1 98,6
780121

Table 2 .2 . Characteristic frequencies below 100 Hz as calculated for the re ­

verberation room of dim ensions 7,85 * 6 ,2 5 x 4 ,9 5 m at the D an ­
ish Technical Highschool

ers have to be measured then the measurements should be performed either

in the open air or in an anechoic chamber.

The three types of normal modes w hich occur in the room are:

1. Axial modes — The component waves move parallel to

an axis (one dimensional, tw o of the indices nx ,n y and
n2 are equal to zero). These are the modes w hich give
rise to flutter echoes. The pressure pattern for an axial
mode is shown in Fig.2.6.

41
2. Tangential modes — The component waves are tangen­
tial to one pair of surfaces but are reflected from the
other tw o pairs (two dimensional, one of the indices
nx,ny and nz is equal to zero).

780782

Fig. 2. 7. Tangential m ode

3. Oblique modes — The component waves are oblique and

therefore impinge on all six walls (three dimensional,
all the indices nx,ny and nz have fin ite values).

42
Fig. 2.8. Sound pressure contours on a section through a rectangular room
fo r the axial mode (2,0,0)

Fig.2.9. Tangential mode (1,1,0)

43
 Fig. 2 .1 0 . Tangential mode (2,1,0)

If one plots the sound pressure contours for various modes of vibration
(Figs.2.8, 2 .9 and 2.10) then one can see that for every mode the sound
pressure is a maximum in the corners of the room. Furthermore, at the geo­
metric centre of the room, only one eighth of the modes have a finite sound
pressure. Therefore, if a monopole sound source be placed in the corner of
such a room, it is possible to excite every mode of vibration to its fullest ext­
ent. Sim ilarly, if a microphone be placed in a corner of the room, it w ill mea­
sure the peak sound pressure for every mode that has been excited. How­
ever, when the monopole source is placed in a position w here a particular
mode has a pressure node, then this mode of vibration w ill be excited only
weakly if at all.

Each of the characterisitc frequencies, f, given by eqn.2.4 may be consid­

ered as a vector in frequency space having the co-ordinates nxc / 2 l x ,
nyc / 2 ly and nzc / 2 lz. A normal mode of vibration is represented by a point
in this frequency space (Fig.2.11). The length of the line joining the point to
the origin is a measure of the characteristic frequency of this particular
mode. A ll the normal modes of a particular charactertistic frequency f and be­
low are included w ithin the octant of frequency space between the positive
axes f x, f y and f z and the spherical surface of radius f. The number and types
of modes having characteristic frequencies lying w ithin a particular frequency
interval can be determined by counting the number of characteristic points on

44
 Fig. 2.11. Characteristic points in frequency space

a frequency space diagram. As this is a tedious task, a num ber of equations

have been derived which yield approximate values for such counts. The num ­
ber of normal modes, N, below a particular frequency, f, is given by

.. 4 n V f3 n S f2 Lf
<2 '5 ’

where V = volume of the room (m3)

S = 2(IxIy + IyIz +lxlz), the total surface area of the room (m2)
L = 4<lx + lY +|2), the sum of the lengths of all the edges (m)
N = number of normal modes

For the reverberation room of dimensions 7 ,8 5 x 6,25 * 4 ,9 5 m mentioned

previously the number of normal modes below the frequency of 100 Hz as cal­
culated from eqn. 2.5 is 4 4 ,8 which is in good agreement w ith the exact
value of 44.

On differentiating eqn. 2 .5 w ith respect to f one obtains an expression

which yields an approximate value for the number of modes, AN, in the band
of frequencies, Af, which is centred on f.

A /V = ( 2 . 6)
m 3

For the reverberation room described above, the number of modes in the
third octave band (i.e. A f is 23% of f) centred on 100 Hz as calculated from

45
eqn.2.6 is 25. Counting the number of modes from Table 2 .2 yields the num ­
ber 28. For the same room, the number of modes predicted by eqn.2.6 in the
third octave band centred on 1000 Hz is over 18000. This means that there
is a fa irly even modal distribution at these higher frequencies and that the
spacing between the characteristic frequencies is so close that specific reson­
ances may be neglected.

Fig.6 .2 8 shows the resonances of a room of dim ensions 2,5 x 3 * 7 m

when excited by a pure tone from a loudspeaker, the electrical power d e li­
vered to the loudspeaker being kept constant. The mode numbers are w ritten
above the corresponding resonance peaks.

If the room possesses w alls which are capable of absorbing sound energy
then the effect of this damping on the standing waves in the room can be in ­
vestigated by introducing a damping term , e~Px , into the undamped standing
wave equation (eqn. A .8) i.e. by replacing jw by jw —/3.

It is theoretically possible to apply boundary conditions at the various w all

surfaces and thus determ ine the characteristic frequencies of the normal
modes. Due to the complexity of the problem, however, at present a solution
can only be obtained for particularly simple configurations of sound absorbing
material. The mathematical argument for the case of damped vibration is s im i­
lar to that for the undamped case except th a t the boundary conditions in the
damped case depend on the absorbing characteristics of the walls. These
characteristics are best described by using the concept of the normal specific
acoustic impedance, zn, w hich is the ratio of the pressure to the normal air
velocity at the surface of the wall (see definition Ch.1). This quantity is both
more fundam ental and more readily measured than the absorption coefficient
of a surface. The normal specific acoustic impedance of a material relative to
air is given by

z„ = (rn + j x n)p c (2.7)

For many materials that are used for w a ll surfaces rn» x n and rn» 1 . For
these particular conditions one finds that the characteristic frequencies of the
damped and the undamped waves are the same.

The damping of the various types of waves depends on the positioning of

the absorbing material. Absorbing material is most effective when positioned
in the corners of a room because it is here that every normal mode has a
pressure maximum.

2.8. PRINCIPLES FOR D ESIG N OF R O O M S AND A U DITO RIA

The size of a room or auditorium and the amount of money to be spent on

46
the construction are tw o factors which are usually decided upon before a spe­
cialist is consulted to assist in designing the acoustics of the building. The
principle acoustical design factor is the reverberation time. Opinions vary con­
siderably as to what is the optimum value of reverberation tim e for an auditor­
ium intended for a particular use. Generally though, one can say that the re­
verberation times for speech and recorded music should be as short as possi­
ble, as one is only interested in the direct and not the reverberant sound; for
light music they should be short and for concert and church music they
should be long. By measuring the reverberation times in auditoria which are
considered to possess good acoustic qualities, one can arrive at a relationship
between the "optim um '' reverberation tim e for a particular use and the vo­
lume of the auditorium (Fig.2.1 2). It should be noted that these curves are in ­
tended only as a guide and that the amount of scatter about these mean
curves is large.

Room volum e in m 3
771205

Fig. 2.12. Typical variation o f reverberation tim e w ith volume for auditoria
considered to have good acoustical properties

47
 The quality of the acoustics of an auditorium , however, does not only de­
pend on the reverberation tim e but also on the shape and size of the enclo­
sure, the positioninq of the sound absorbing material and the positioning of
the sound source and the audience.

The acoustical defects w hich can arise due to the size and shape of the
room are echoes, dead spots and flutter.

An echo occurs when a strong reflection of an original impulse is heard by

a listener after an interval of greater than 0 ,0 5 sec since he heard the o rig i­
nal impulse. The ear functions in such a w ay that when the interval between
the original and the reflected sound is less than 0 ,0 5 sec, then the reflected
sound is not recognised as a separate echo but adds to the apparent strength
of the original sound. This phenomenon is known as the Haas effect.

Dead spots can occur at positions in auditoria which are far from reflecting
surfaces and which receive sound only after it has passed over a particularly
absorbing surface. Dead spots can occur, for example, at the back of cinemas
and theatres where the sound travels at almost grazing incidence over the au­
dience.

The phenomenon of flu tte r occurs w hen both the source and the listener
are between a pair of parallel, hard, surfaces and other nearby surfaces are
fairly absorbent. Sound emitted by the source w ill tend to be "trapped” be­
tween the reflecting surfaces and w ill oscillate back and forth and decay rela­
tively slowly. The listener w ill perceive this oscillating energy as a "flu tte r­
ing" of the sound.

2.9. DESIGN OF R O O M S FOR SPEECH

W hen designing a room for speech, the most im portant criterion is that
the speaker should be distinctly and readily heard by all members of the audi­
ence. A quantitative measure of the degree of clarity at various positions in
the room can be obtained by articulation tests. These tests consist of reading
aloud from the speaker's platform a list of monosyllabic nonsensical words.
The people in the audience then w rite down w ha t they think they have
heard. An analysis of the percentages of consonants and vowels w hich were
heard correctly is then performed to give "the percentage articulation index"
(P.S.A ). Normal, connected speech can be understood even if some of the
syllables are unintelligible. This is due to the fact th a t the listener can de­
duce the meaning from the context of the sentence. Even under perfect listen­
ing conditions, the maximum value of P.S.A. obtained is normally about 95%
due to unavoidable errors. A P.S.A. of 80% enables the audience to under­
stand every sentence w ithout undue effort. In a room where the P.S.A. is

48
about 75%, the listener has to concentrate to understand what is said while
below 65% the intelligibility is too poor.

Assuming there is no speech am plification system present, there are basi­

cally four factors w hich affect the clarity of speech in a room. Firstly, there is
the background noise level w hich can mask the desired sound. This level
should be kept below 30dB A . Secondly, there is the sound pressure level pro­
duced at the listener's ear by the speaker. This depends on the distance be­
tween the listener and the speaker, the volume of the room and the nature of
the speaker's surroundings. Thirdly, there is the reverberation tim e. In nor­
mal speech the syllables follow each other with rapidity. Unless each syllable
decays fairly quickly it w ill tend to mask those fo llo w in g. Therefore the reverb­
eration time m ust not be too long. On the other hand, if the reverberation
tim e tends to zero then the absence of reflected components severely res­
tricts the size of a room in w hich the speech w ill be heard at a sufficiently
high intensity for good intelligibility. Lastly, there is the room shape, although
providing this has been designed to avoid echoes and dead spots and each
member of the audience has a good view of the speaker then the articulation
should not be affected.

A compromise has therefore to be made between a loss of definition due to

excessive reflected sound and a loss of sound intensity due to excessive ab­
sorption at the room's boundaries.

In practice a speaker may adapt himself to a particular hall by speaking

more slowly and loudly when addressing a large audience but there are limits
to the amount of accommodation possible and a speech amplification system
may have to be employed.

Once one knows the volume th a t the room is to be and the required reverb­
eration time, one is in a position to decide how m uch sound absorbent treat­
ment is necessary. The positioning of the absorbing material w ith in the room
w ill depend on the particular circumstances but in general the greater part of
the material should be placed at the end of the room opposite to the speaker
and mounted on those surfaces w hich are likely to give rise to troublesome
reflections. Absorbent materials of a resilent nature such as perforated
boards or wooden slats backed w ith mineral w ool should be used on walls
w here they are liable to suffer damage or wear. Less robust m aterials, such
as acoustic fibre tiles, should only be used out of hand's reach.

There are certain rooms w hich are used for speech which require special
attention such as rooms used fo r debate, theatres, lawcourts and m ultipur­
pose halls e.g. school halls, com m unity halls etc.

In debating chambers each person must have good sight lines to all the oth­
ers present and the intelligibility must be acceptable at all positions in the

49
chamber. In theatres it is important that the natural qualities of the perform­
ers' voices be maintained. However this requirement of naturalness does not
apply to public meeting halls. Rooms used for both speech and m usic have
conflicting requirements (i.e. a short reverberation tim e to render speech in­
telligible and a long reverberation tim e to give depth to music) so th a t a com­
promise has to be decided upon. If a sound am plification system is to be used
then this compromise is not such a critical decision. The reverberation time
can be made long enough for music and the sound amplifying system used to
overcome the long reverberation tim e. The subjective reverberation tim e of
an auditorium can be adjusted to s u it the programme by employing electroa-
coustical techniques, for example, the assisted resonance technique.

Fig. 2.14. Helm holtz resonator used in the assisted resonance technique

50
 This technique increases the reverberation time by prolonging the dying
away of the room modes or resonances in the auditorium . For a particular
mode whose reverberation has to be "assisted" a Helmholtz resonator is posi­
tioned in the auditorium w here there is a pressure maximum for this particu­
lar frequency (Fig.2.13). The resonator is then tuned to th is frequency by ad­
justing the volume of the resonator. A microphone placed inside the cavity of
the resonator detects the selected frequency and feeds the signal via an am ­
plifier to a loudspeaker w hich then em its acoustical energy at this frequency
back into the auditorium . This "topping u p " with acoustical energy, compen­
sates to a controllable extent for the sound energy absorbed by the audience
and the room surfaces. The rate of decay of sound is th us slowed down and
the reverberation tim e is therefore lengthened. To deal w ith a range of fre ­
quencies, many microphone-loudspeaker systems have to be employed. The
Royal Festival Hall in London, for example, has 172 such channels covering
the frequency range from about 60 Hz to 700 Hz. The resonators can be
placed anywhere in the auditorium except close to the source of the music be­
cause it is the reverberant field, not the direct field, th a t the resonators must
respond to. The loudspeakers can be placed anywhere in the auditorium (ex­
cept very close to the listeners) because w ith these tuned circuits it is impos­
sible to tell where the extra sound energy is coming from . This system has
been in use in the Royal Festival Hall, London, since 1 9 6 2 .

2.10. DESIGN OF R O O M S FOR M U S IC

It is much more d ifficu lt to state criteria for good listening conditions for
music than for speech because aesthetic and emotional judgements are in ­
volved. The criteria are almost totally subjective making them very d ifficult to
define and often impossible to measure. The design of rooms for music is
therefore as much an art as a science.

The notable survey by L.L.Beranek o f sixty concert halls situated all over
the w orld enabled 1 8 acoustical factors to be listed in order of their im port­
ance. From the results of his researches, Beranek developed a system of ra t­
ing concert halls in term s of these 1 8 factors. These researches are de­
scribed in the book "M u sic, acoustics and architecture" by L.L.Beranek pu­
blished by John W iley and Sons, New York, 1 962.

Some of the more im portant factors w ill be discussed here.

Loudness

The music in the room must be sufficiently loud. As the sound energy avail­
able from a musical instrum ent is lim ited, this therefore sets a lim it on the
size of the auditorium.

51
Reverberation

There should be sufficient reverberant sound. The amount of reverberant

sound required depends upon the nature of the music. M usic from the bar­
oque period or chamber music requires a short reverberation tim e, whereas
music from the classical period such as orchestral works by Tchaikovsky and
Wagner demand longer reverberation times.

Definition

The music should possess definition or clarity. This quality is basically the
ability of the listener to differentiate between the different instrum ents in an
orchestra and between the different musical sounds. D efinition contradicts
the requirement for sufficient reverberant sound.

Fullness of tone

This quality describes the blending effect that reverberation has on succes­
sive notes and chords when heard in a room. Fullness depends mainly on the
reverberation time. The longer the reverberation tim e (within reason) the bet­
ter the chance of obtaining adeuate fullness.

Definition and fullness are interelated. As for speech, d efinition depends

on listeners receiving the direct sound and firs t reflections of it arriving not
more than 35 ms later at a strength well above the reverberant sound level.
On the other hand fullness appears to depend mainly on having plenty of re­
verberant sound. Thus it is likely that definition w ill suffer if the reverberation
is made great enough for maximum fullness.

No obvious faults

There should not be any obvious faults in the music room such as notice­
able echoes and dead spots.

Intimacy (or presence)

This quality relates to the sense of being enclosed in a space and enve­
loped in the sound field. Sound must be reflected from many surfaces to the
listener from many directions so that he w ill sense the space in which he is
sitting. Intimacy depends upon the time delay between the direct sound and
the first reflection. It is found th a t in narrow halls having a very small initial
time delay of about 15 ms the sense of intim acy in the centre of the main
floor is extremely high.

52
Musicians' criteria

The musicians themselves have tw o important criteria to be satisfied.

Firstly, the room should respond to th e ir instruments. This means that some
sound should be reflected back to them from the listening area but care is
needed to avoid echoes. Secondly, the musicians should be able to hear each
other. Reflecting surfaces around the orchestra w ill help to accomplish this.

The ultimate test fo r a room or hall intended for m usic is of course to listen
to a live performance. An experienced listener moving about the hall during
the test performance should be able to detect the presence of any faults. It is
then the job of the acoustical consultant to suggest remedial measures.

2 .1 1 . REFLECTORS, ABSORBERS A N D RESONATORS

Sound reflectors and sound absorbers are often used to produce the de­
sired acoustical conditions in rooms and auditoria. Some of the properties
and applications of sound reflectors and absorbers w ill be discussed here.

Sound reflectors

The ratio of the am ount of reflected to incident sound at the boundary be­
tween two media depends on their relative acoustical impedances. For a
material to be a reflector of sound its impedance w ill therefore have to be d if­
ferent to that of air. As impedance is equal to the product of density and the
speed of sound in medium, reflectors in an airborne sound field should be
massive. Geometrical reflection of the incident sound wave can only be as­
sumed when the reflecting surface is large compared to the wavelength of
the incident energy, thus practical reflectors in a room w ill have a definite
low frequency cut off below which they act as diffusers. Generally speaking,
a panel with a m inim um dimension of about 30 tim es the wavelength of the
incident sound acts as a reflector; w he n the minimum dimension is about 10
tim es the wavelength some diffraction occurs, w hile at less than 5 tim es the
wavelength the incident energy is diffracted.

M any auditoria employ suspended overhead reflectors (e.g. the A lbert Hall,
London) in an attem pt to combine a sufficiently large volume to obtain sui­
table decay times for music and short path reflections to produce good d e fin i­
tion.

Sound absorbers

Every surface of a room and every object w ithin it w ill absorb sound to
some degree. Hard smooth objects such as walls and tiles w ill absorb fa r less

53
sound energy than soft, porous materials such as carpets, upholstery and peo­
ple. The porous sound absorbing materials used for the control of acoustical
environments (e.g. mineral w ool, glass fibre) are characterised by a number
of variables, i.e. porosity, flow resistance and structure factor.

A sound wave impinging on th is type of absorber causes the air in the con­
nected pores to vibrate but as the movement of the air particles is restricted
by the flow resistance of the m aterial, some of the sound energy is dissipated
as heat. The amplitude of vibration of the air particles is progressively
damped by friction against the pore walls. This acts as an acoustical resist­
ance depending chiefly on the resistance of the material to direct a ir flow.

The effective porosity of the material (i.e. the volume of pores connected to
external air compared to the total volume of the material) governs the amount
of airborne sound energy that may enter and be subjected to attenuation.
However the structure factor is also important since air in blind alleys and
pores which run parallel to the surface w ill be little affected by air flo w in the
main pores and have negligible effect on the absorptive effect.

In practice the actual absorption chracteristics of many of these sound ab­

sorbers depends on the method of mounting. M axim um sound absorption w ill
occur in a porous absorber when the particle velocity in the absorber is at a
maximum. In a sound wave w hich is incident on a rigid w all, the maximum

Frequency 162354

Fig. 2.15. Absorption characteristics of an absorptive material m ounted 9 cm

from a solid wall

54
 Frequency 162353

Fig. 2 .1 6 . Dependence o f the absorption of a porous m aterial on the thickness

of the m ate ria l
particle velocity is to be found at 1 / 4 wavelength from the wall. Thus in or­
der to absorb low frequency energy the absorber m ust either be very thick or
else be mounted some distance in fro n t of the w all. The absorption character­
istics of a porous m aterial mounted at a distance of 9 cm from a solid w all is
shown in Fig.2 .1 5 . The greatest absorption occurs at frequencies for which
the distance between the wall and the absorptive material corresponds to a
1 /4 and a 3 /4 wavelength of these frequencies. F ig.2.1 6 shows the depen­
dence of the absorption of a porous material on the thickness of the material.

Panel sound absorbers

If a plate or panel is mounted in front of a rigid w all, the arrangement be­

haves in the same way as a spring-mass system, the plate being the mass
and the enclosed air being the spring. When a sound wave impinges on the
system, the system w ill tend to be set into vibration. The maximum transfer
of energy occurs w hen the frequency of the incident sound energy is the
same as the resonant frequency of the system.

Since the panel possesses inertia and damping, some of the sound energy
is converted into mechanical energy and dissipated as heat, therefore sound
absorption occurs. However since the panel itself vibrates it w ill act as a
sound radiator so th a t it is rare to find such a system w ith an effective absorp­

55
tion coefficient greater than 0 ,5 . The resonant frequency of such a system
can be calculated from

6000
( 2 . 8)
res “

w here m = mass of plate in kilograms per square meter

d = distance between plate and wall in metres

Such panel resonators are useful at mid- and low-frequencies (Fig.2.1 7). The
absorption of sound energy falls off rapidly at frequencies above the resonant
frequency. Further damping may be obtained by introducing damping material
into the air volume between the w all and the panel. This broadens the range
of frequencies for which the resonator is active.

Resonator absorbers

The simplest type of resonant absorber is called a Helmholtz resonator.

This consists of a volume of air contained w ithin a cavity connected to the air
in the room by a small opening known as the neck (Fig.2.18). W hen a sound
wave impinges on the aperture of the neck, the air in the neck w ill be set
into vibration and the air in the cavity w ill undergo periodic compression and
rarefaction. The friction between the resulting am plified motion of th e air par­
ticles in the neck and the neck itself causes sound energy to be absorbed.
The absorption of the undamped resonator falls off very rapidly at frequencies
above and below the resonant frequency. Some energy is transferred into the
cavity and this energy is reradiated after the incident sound has ceased. Thus
the resonator may prolong the reverberation time of a room. The resonant fre­
quency of an undamped resonator may be evaluated from

(2.9)

56
where c = velocity of sound (m .s.~1 )
S = cross sectional area of neck ( m 2 )
I = length of neck (m)
V = volume of cavity (m 3 )

______________________________________ 171029

Fig. 2 .1 8 . Helmholtz resonator and the equivalent m echanical analogy

If the resonator be damped i.e. by lining the cavity and neck w ith a porous
sound absorbing material then the resonator w ill be effective over a w ider
bandwidth although its maximum absorption at resonance w ill be reduced
(Fig.2.1 9). Helmholtz resonators can be designed to provide absorption at any
point in the frequency scale but owing to their sharp tuning they are not o f­
ten used for general acoustic treatment w here a major change is required but
only w here a particularly long reverberation is experienced at a single fre-

Fig.2.19. Typical sound absorption characteristics of a Helm holtz resonator

w ith and w ithout damping

57
quency such as that due to a normal mode and it is desired to reduce this
w itho ut greatly affecting the average reverberation.

Helmholtz resonators are most efficient at low frequencies.

Perforated panel absorbers

A much commoner application of the Helmholtz resonator principle for

sound absorption is to be found in the acoustic panel. This consists of a
panel or plate w ith a drilled or punched pattern of holes, mounted in such a
way so as to enclose an air space between itself and the w a ll (Fig.2.20). The
slots or holes form the necks of the resonators and the portion of the air
space behind each hole form s the cavity of the resonator. Usually there is no
need to divide the separate resonator cavities from one another by partitions.
As for the single resonator, the resonant frequency of the m ultiple resonator
is determined by the dim ensions of the neck and the cavity but the m ultiple
resonator is not so selective in its absorption. In practical acoustic panels, a
damping material such as m ineral wool or glass fibre is inserted into the air
space thus increasing the effectiveness of the absorption above and below
the resonant frequency (Fig.2.21).

Fig. 2 .20 . Perforated panel absorber seen as an assembly o f Helmholtz reso­

nators

For panels w ith regularly spaced holes of the same diameter such resona­
tor has the same resonant frequency. It is possible to alter the spacing and di-

58
ameter of the holes to produce a panel w ith the desired absorption character­
istics.

A commercial type of perforated panel absorber is the "Kulihat . This is a

conical absorber, resembling a coolie's hat, consisting of two or three perfor­
ated aluminium sectors held together by steel clips. The inside of the Kulihat
is lined w ith m ineral wool (Fig.2 .2 2 a) and b). The absorption characteristics
for tw o and three section Kulihats are shown in F ig.2.23.

Fig. 2.22. The "K u lih a t" perforated sound absorber

a) three sectors b) tw o sectors

Sabines m 2

F reque ncy in Hz __________________ l 7 , 035

Fig. 2.23. Sound absorption characteristics fo r "K uliha t" absorbers

a) three sectors b) two sectors
59
 Functional absorbers

In some rooms and halls there are insufficient surfaces available for the
m ounting of the necessary sound absorbing m aterial. In this situation func­
tional absorbers can be employed. These are three dimensional units of
sound absorbing m aterial suspended freely in the room some distance from
the room boundaries. Since sound energy can im pinge on all sides of these
units, their efficiency is high. This type of absorber is particularly useful in
workshops, gymnasia, swimming pools etc.

60
 3. ACOUSTICS OF BUILDINGS

3 .1 . INTRODUCTION

So far in this booklet, only the behaviour of sound w ith in rooms and enclo­
sures has been discussed. Now attention w ill be focussed on the trans­
mission of sound from one room to another and the sound insulation propert­
ies of building elem ents i.e. w alls, floors, doors and windows. Sound insula­
tion is especially pertinent when designing m ulti-room ed buildings such as
blocks of flats, hospitals, schools etc., where practically all intrusive noise is
considered to be highly undesirable.

3 .2 . SOUND GENERATION M ECHANICS

Sound can propagate throughout a building e ithe r via the air or via the
building's structure (Fig.3.1.). Sound generation mechanisms can therefore
be divided into tw o general groups.

Fig. 3 .1 . Intrusive noise due

a) airborne noise b)

One group consists of those sources which generate sound directly into the
air such as the voice, loudspeakers etc. Insulation against such sound is
called airborne sound insulation.

61
 The other group consists of those sources w hich act directly on the struc­
tu re of the building usually by m eans of impact or vibrating equipm ent. Trans­
m ission of the sound is then through and from the structure. Examples are
footsteps, noisy plumbing installations and slamming doors. In fact, this type
of noise is really a combination of both airborne and impact noise because
the impacts w ill produce airborne noise and this airborne noise w ill be trans­
m itted. However, in nearly all cases the noise produced in the receiving room
by the transmission of the impact noise w ill predominate. Insulation against
such sound is called impact sound insulation.

3 .3 . SOUND INSULATION

W hen discussing the insulation th a t exists between two rooms against air­
borne or impact sound, it is usual to consider one room as the source room
and the other as the receiving room. The basic problem is to determine
(a n d /o r reduce) the sound pressure levels produced in the receiving room
due to a source acting in the other room. The sim plest case to consider is the
one where the tw o rooms have one common dividing element i.e. a w all or a
flo o r/ce ilin g .

3 .4 . AIRBORNE SO UND INSULATION

A noise source operating in one room w ill produce a reverberant sound

field which impinges on all surfaces of the room. The sound energy incident
upon the dividing w all w ill depend on the sound power output of the source
and the total sound absorption in the room. This incident sound energy w ill
be partly reflected back into the room and partly "absorbed" by the w all, the
ratio depending on the w all's acoustical impedance compared to th a t of air.
Of the absorbed energy some w ill be dissipated as heat, and the rest w ill be
propagated through the wall to the boundary w ith the receiving room. At this
boundary, the relative impedances of the wall m aterial and the air w ill again
determ ine the percentage of energy transmitted into the receiving room. The
overall effect is that the entire w a ll is forced into vibration by pressure flu ctu ­
ations of the incident sound waves. (Usually the vibration of the w a ll can only
be detected w ith the aid of an accelerometer or sim ilar transducer but with
particularly intense sound waves one can feel the w all vibrating.) The vibrat­
ing w all, acting in exactly the same manner as a loudspeaker, then radiates
acoustical energy into the adjoining room.

The amount of radiation from the w all and hence the sound insulation pro­
vided by the w all depends on the frequency of the sound, the construction
and material of the w all and above all on its w eight. As one can im agine, the
more massive the w all the more d ifficu lt it is for the sound waves to set it
into vibration.

62
 The sound insulation characteristic of a wall is usually expressed in term s
of sound reduction index, R, (referred to as the transm ission loss in the
United States of America) expressed by

/? = 70/o<710( ^ ) dB (3.1)

w here W , = sound power incident on the wall

W 2 = sound power transmitted through the w all

A typical sound insulation curve for a 15 cm concrete w all is shown in

Fig.3.2. For example, if the sound reduction index of a particular w all were
40 dB then the incident sound energy would be reduced by 40 dB in tra n s ­
mission through it. Thus only one ten thousandth of the incident energy
would be transm itted. The larger the value for the sound reduction index of a
w all, the greater the sound insulation it provides.

Frequency, Hz 7 7 , t5 t

Fig. 3.2. Sound reduction index as a function o f frequency for a 15 cm con­

crete w all as measured in a laboratory and as m easured in situ.

The sound reduction index depends on the angle of incidence of the im ping­
ing sound. If the sound fields in the tw o rooms are diffuse and providing the
sound is transm itted only through the dividing wall then the sound reduction
index may be evaluated from

63
 (3.2)

where L , = average sound pressure level in the source room

L2 = average sound pressure level in the receiving room
S = area of the dividing wall
A = equivalent absorption area of th e receiving room determined
from reverberation measurements

Except in specially designed laboratory transm ission suites, there is practi­

cally always a certain amount of indirect or flanking transm ission. Various
possible transmission paths between adjacent rooms are shown in Fig.3.3.
Besides the structure borne paths shown there could also be transmission
through airducts, leaks around doors etc.

To account for flanking transm ission the quantity known as the apparent
sound reduction index, R', is used. This is defined as

R '- W lo g , „ ( £ - ;) (3.3)

64
w here W-| = sound power incident on the wall
W 3 = total sound power transm itted into the receiving room

The apparent sound reduction index may be evaluated from an expression of

the form of eqn.3.2 i.e.

dB (3.4)

In a practical situation (e.g. a block of flats) it is the apparent rather than

the absolute sound reduction index th a t one is interested in.

3 .5 . IM PACT SO UN D INSULATION

Sources of impact sound (e.g. footsteps) act directly on the structure of the
building, causing the structure to vibrate and to radiate acoustical energy into
the receiving room.

The impact insulation characteristic of a floor is usually expressed in terms

of the impact sound pressure level, L j , which is the average sound pressure
in a specified frequency band in the receiving room when the floor under test
is excited by a standardized impact source.

To allow for the effect of the absorption of sound which occurs in the re­
ceiving room a correction is applied to the impact sound pressure level yield­
ing the quantity known as the normalized impact sound pressure level de­
fined by

dB (3.5)

w here Ln = normalized impact sound pressure level

Lj = impact sound pressure level
A0 = 10 m2 the reference absorption area
A = measured equivalent absorption area of the receiving room

Except in specially designed laboratory transm ission suites, flanking trans­

mission of the impact sound w ill nearly always be present. A typical norm al­
ized impact sound pressure level curve for a solid concrete floor is shown in
Fig.3.4. Among the ingenious methods devised for providing a standard im ­
pact are a person of a specified w eight walking at a standard pace and w ear­
ing standard shoes for exciting floors; and ball bearings shot from a spring
loaded gun for exciting walls and ceilings.

65
 60

20 --,-- --,--,--,--,-- -- -- -- ,-- -- -- -- --

100
1
200
1
500
1 1
1000
1 1 1
2000
1

Frequency, Hz 771149

Fig. 3.4. Typical norm alised impact sound pressure levels measured directly
below a concrete slab floor using a standard tapping m achine as an
impact source
a) bare concrete floor
b) concrete flo or covered w ith a rug

The most common method used at present for the excitation of floors is the
standard tapping machine. This machine contains five metal hammers which
are lifted by means of an electric motor then allowed to fall and strike the
floor one after the other.

The improvement of the impact sound insulation (i.e. the reduction of the
impact sound pressure level) is the difference betw een the average sound
pressure levels in the receiving room before and after some treatm ent of the
floor e.g. the installation of a floor covering, a floating floor or a false ceiling.

Material dB/30 m
CO
1

Iron
0

Brickwork 0 , 5 - 4,0
CD
CD

Concrete
I

Wood 1,5 - 10,0

780119

Table 3.1. A ttenuation of longitudinal waves in building m aterials

66
 The attenuation of noise borne in building materials is usually quite small
(Table 3.1). Therefore, machines and other sources w hich are likely to pro­
duce structure borne noise should be isolated from the main building struc­
ture wherever it is possible.

3.6. AIRBORNE S O U N D REDUCTION INDEX FOR A SOLID H O M O ­

GENEOUS IM PERVIO US WALL

The relative importance of the different mechanisms of sound transmission

through a solid varies throughout the audio frequency range. A solid w all pos­
sesses the qualities of stiffness and mass and can thus exhibit resonance
and mode effects. A t low frequencies transmission depends mainly on the
stiffness of the w all i.e. damping and mass are unimportant. At slightly
higher frequencies, the resonances of the wall control its behaviour. A t a fre ­
quency of about tw ice the lowest resonance frequency, the walls tends to be­
have as an assembly of small masses and is said to be mass controlled. From
theoretical considerations of the transfer of energy from randomly incident
sound waves to the particles of the w all and then into the air on the other
side, the relationship known as the Mass Law can be derived. The Mass Law
can be expressed as

R = 20 lo g :0 ( fM ) - 47 dB (3.6)

where f = frequency of the incident sound, Hz

M = surface density of the w all in kg /m 2

This relationship gives the theoretical maximum sound reduction index for
random incidence.The Mass Law should only be used to give an approximate
guide to the insulation obtainable. In practice, the sound insulation obtained
is always a few dB less than the theoretical maximum. In the mass controlled
region the sound reduction index increases at a rate of about 6 dB for each
doubling of frequency i.e. by 6 dB per octave, and by about 6 dB for each dou­
bling of surface density, which means, for a particular m aterial, for each dou­
bling of thickness. The mass controlled region extends up to the critical fre ­
quency. This is the frequency at which the wave length of the bending waves
in the w all is the same as the wave length of the radiated sound in the receiv­
ing room or in other words, the low est frequency capable of exciting the
coincidence effect (see the following section). Above the critical frequency
the stiffness of the w a ll again plays an important role.

The various regions are shown in the idealized sound reduction index curve
in Fig.3.5.

67
 Critical Frequency
Frequency, Hz 7 7119 8

Fig. 3.5. Idealised sound reduction index curve for various degrees o f damp­
ing

3.7. C O IN C ID EN C E EFFECT

In air sound is propagated by longitudinal waves at a velocity w hich is the

same for all frequencies. However in solid structures such as a w all, sound
can be propagated by longitudinal, transverse and bending waves. The most
important from the point of view of building acoustics are the bending waves.
These waves are associated w ith large transverse displacements which
means that they can readily couple to longitudinal waves in the surrounding
air.

Bending waves of different frequencies travel at different velocities, the ve­

locity increasing w ith frequency. This means th a t for every frequency above a
certain critical frequency there is an angle of incidence for w hich the wave­
length of the bending wave can become equal to the wavelength in air pro­
jected onto the wall. This is the condition known as coincidence (see
Fig.3 .6 .).

When coincidence occurs it gives rise to a more efficient transfer of energy

from the air to the wall to the air on the other side of the w all. Thus the effec­
tive insulation of the w all is lowered producing the "coincidence dip" in the
insulation curve (Fig.3.5).

68
 The condition for coincidence to occur is

;
sin d = ~ (3.7)

w here A = wavelength of sound in air

= wavelength of sound in the wall

Clearly,if the wavelength of the sound in air is greater than the wave­
length of the sound in the wall, no coincidence can occur because the sine
cannot be greater than 1. When the incident sound has a fixed frequency,
the angle at which coincidence occurs is defined as the coincidence angle.
W hen the angle is fixed then the frequency at which coincidence occurs is de­
fined as the coincidence frequency.

The critical frequency is defined as the lowest frequency at which coinci­

dence occurs i.e. that frequency for w hich A = , or c = c 8 .

w here c = velocity of sound in air

cB = velocity of sound inthe w all

In other words the critical frequency is the lowest possible coincidence fre ­
quency and occurs w hen the sound is at grazing incidence to the w all i.e. 9 =
90°.

In many cases the coincidence dip occurs in the frequency range 1 0 0 0 Hz

to 4 0 0 0 Hz, which includes im portant speech frequencies. When specifying
the sound reduction index for a dividing wall it is therefore necessary to de­
fine the insulation required over the w hole frequency range, because a single

69
figure representing the averaging sound reduction index does not show defi­
ciencies due to resonance or coincidence effects.

3 .8 . M ETHODS OF IM PR O V IN G AIRBORNE SO UND INSULATION OF

BUILDING ELEMENTS

Damping

For a building element to have a high sound reduction index over a wide
frequency range, the element (w all, floor, door, window) should possess a
high mass and a low stiffness. An element may have sufficient mass to pro­
vide good insulation but its fu ll potential is not realised because high stiff­
ness narrows the frequency range between resonance and coincidence. It is
not usually possible to reduce the stiffness of an existing element but the ef­
fects of the stiffness can sometimes be reduced by increasing the damping in
the element. Damping is only effective in the frequency ranges w here reson­
ance and coincidence occurs. There is virtually no effect on the sound reduc­
tion index due to damping, in the frequency range where the mass law ap­
plies (see Fig.3.5). The most common method of adding damping to an ele­
ment is to apply a thick layer of a mastic-like m aterial to one side o f the ele­
ment. Providing there is a good bond between the layer and the element a
greater proportion of the energy incident in the element w ill be dissipated in
the layer. M ultilayered "S an dw ich" structures can be built up in th is way.

This type of treatm ent is only effective on elem ents that have inherently a
small amount of damping and low superficial mass. It would be useless for
example to apply a damping layer to a 1 5 cm thick concrete wall although the
performance of metal partitions can be greatly improved by this treatm ent.

Double Leafed Elements

The insulation of a single leaf element can be improved by increasing the

mass of the elem ent but this process can only continue up to a certain point.
For example if a proposed office partition of surface mass 5 0 k g /m 2 is re­
placed by a thick concrete w all of 2 0 0 k g /m 2 then the improvement in the in­
sulation w ill be about 10dB. The increase in cost and weight may be ju stifi­
able in certain cases. If, however, a brick w all exists already and a 10dB im­
provement in the insulation is sought then quadrupling the mass o f the brick
w all would alm ost certainly not be a practical proposition, quite apart from
the fact that at these high values of sound insulation, the flanking w alls usu­
ally constitute the principal transm ission paths. It may at first be thought that
one way of increase the insulation would be to construct a second w all just
behind the first. Thus, if the w a lls were identical and the insulation due to
one wall alone w ere 4 0 dB then the total insulation due to both w alls would

70
be 40 + 40 = 80 dB. U nfortunately in practice th is is just not so. If the walls
were completely separated from each other w ith no common footings or edge
supports and no connecting ties and the air spaces between the w alls were
larger than about a meter then the total insulation would approach the value
of 8 0 dB. In practice though, w alls usually have common supports at the
edges and it is rare to find a double leaf w all (or cavity wall) w ith a cavity
w ider than a fe w centimeters. Consider the transm ission of sound through a
double leaf elem ent such as the one shown in F ig.3.7. There are several pos­
sible transm ission paths through the cavity, through the edge connections
and through the connecting ties. The graph in Fig.3 .8 shows the insulation
obtained when the leaves are entirely unconnected and the effect of the com­
mon connections.

_______________________________________________________________________________________________________ 7 71148

Fig. 3. 7. Transmission paths through a double leaf w a ll

At low frequencies the air between the tw o leaves couples them together
rather like a spring. A resonance frequency, f r , therefore exists which is de­
termined by the mass of the leaves and by the w idth of the cavity. A t the re­
sonance frequency there is a sharp drop in the value of the insulation. In
practice f r should be made less than 100 Hz so that for leaves of low mass a
wide cavity is essential. Above the resonance frequency, the insulation in ­
creases more rapidly w ith frequency than it w ould for a single w all of the
same weight but above about 2 5 0 Hz the problem of cavity resonances arises
which tend to reduce the insulation. These resonances can be considerably
reduced by lining the cavity w ith an absorbent material such as mineral wool
or glass fibre. It is not necessary to fill the w hole cavity. A layer of about 3
cm of absorbent should suffice. An absorbent lining is most effective in light­
w eight constructions and has almost no effect on a heavy masonry or con­
crete wall.

71
 70

Fig. 3.8. Transmission loss

a) theorectical transm ission loss of a double w a ll of mass M
b) typical m easured transm ission loss for a double w all
c) Mass Law for a single w a ll o f mass M

Transmission through common ties and around the common perim eter be­
comes important when an insulation of more than 40 dB is required. For insu­
lation in excess of 50 dB the source and the receiving rooms should be vibra-
tio na lly isolated from each other.

Flanking Transmission

Frequently, the expected insulation of a dividing w a ll is not realised in

practice due to flanking transmission (see Fig.3.3). There are several other
transm ission paths that sound can fo llo w apart from the direct path through
the dividing wall. To obtain the desired insulation none of the flanking paths
should be weaker in insulation than the direct path.

The existance of flanking paths means that in some cases it is pointless to

construct a high quality insulating w a ll without m aking major structural
changes in the other building elements of the room.

Providing the critical frequencies of the dividing w all and the flanking ele­
m ents are low compared to the frequency range of interest then the am ount

72
of energy radiated into the receiving room by the various flanking elements
can be measured and compared w ith the am ount of energy radiated from the
dividing wall by employing the relationship (Ref.2)

W k = pcSkVk2 crk <3 -8 >

where W k = the sound power radiated from elem ent k

Sk = the area of elem ent k
V2 = the average mean square normal surface velocity
crk = the radiation efficiency which has the value of I above the c riti­
cal frequency
pc - the characteristic impedance of air which has the value of 41 5
rayls at standard atmospheric tem perature and pressure

Doors

The insulation provided by a door does not follow the predictions of the
Mass Law for tw o reasons:

a) there are nearly always sm all gaps between the door and the door-frame
through w hich sound can be transmitted

b) the size of the door is very much sm aller than the w all in which it is
placed so that the resonant frequencies of the door occur at much higher
frequencies than in a w all made of the same material.

The insulation provided by a door as estimated by the Mass Law w ill there­
fore always be somewhat higher than that w hich can actually be obtained.

When high insulation is required the edges of the door should be sealed
very carefully w ith gaskets of fe lt or rubber.

Values of insulation greater than those predicted by the Mass Law can be
obtained by the use of double doors. If the tw o doors are separated by a short
passageway, sound absorptive material applied to the walls and ceiling in the
passageway w ill further improve the insulation. For double doors separated
by at least 8 cm the average insulation is estimated to be at least 5dB grea­
ter than the Mass Law value based on a mass equal to the sum of the
masses of the tw o doors. The seal around the doors must be as carefully
made as for the single door. Seals can be provided by compressible gaskets,
drop-tongue draught (and noise) excluders and covered key holes.

73
Outer Walls and Windows

M any of the noises which disturb people w ithin buildings originate out of
doors. The noise produced by such sources as air and road traffic and indus­
trial noise enter the building via the outer walls, the w indow s and the roofs.
M any modern buildings have roofs made from concrete or comparable heavy
m aterials so these do not pose such a great problem as the outer w alls and
w indow s.

The insulation of the outer facade is usually but not always determined by
the insulation of the windows. When a high degree of insulation is required,
it is essential that fixed windows be used which means that a mechanical
ventilation system is necessary.

The insulation provided by a w indow is more difficult to estimate than that

of a w all because the insulation is more dependent on the window 's d im en ­
sions and the coincidence effect plays an important role. A rough estimate of
the insulation can be obtained from the Mass Law.

The insulation curve of a single glazed window is typically marked by a

deep trough in the mid-frequency range; the range in w hich the ear is most
sensitive. This is not so detrimental to the insulation w hen it is the noise
from tra ffic that has to be excluded because the low frequencies predom i­
nate. However, the short comings of a w indow 's insulation are more notice­
able w hen the source of sound is a je t aircraft because of the w indow 's poor
insulation at high frequencies.

As w ith walls and doors,an improvement in the insulation of a w indow can

be obtained by using a double leaf construction i.e. tw o panes of glass separ­
ated by an air gap. The behaviour of such a construction is analogous to a
mass-spring-mass system, as the masses of the panes of glass are relatively
low, the stiffness of the spring (i.e. the thickness of the air gap) should be
large, in order to make the resonant frequency sufficiently low. The double
glazing w hich is used for the purpose of therm al insulation usually has a very
narrow cavity (about 1,0 to 1,2 cm) so th a t the masses are coupled through a
stiff spring. Consequently, the resonant frequency is situated at approxi­
mately 3 0 0 Hz and thus the sound insulation in this frequency region is disap­
pointingly small. Double glazing (i.e. a double window) can be made an e ffi­
cient sound insulator providing the w id th of the cavity is at least 3 inches
(75 mm). A further improvement in the insulation is obtained by placing
sound absorbing material on the sides of the window fram e w ith in the cavity.

Floor-Ceiling Elements

The same principles apply for the airborne sound transm ission of flo o r-ce il­

74
ing elements as for walls. However, special consideration m ust be given to
the control of impact noise to which the floor is subjected because a design
that is a good insulator for airborne sound may be unacceptable from the
point of view of the transmission of impact noise e.g. a concrete slab floor.

An obvious solution to the problem of impact insulation is to reduce the im ­

pact effect on the main structure by covering the floor w ith a resilient layer
such as carpeting or rubber tiling . The action of the resilient layer is to cush­
ion the blow of the impact and thus to reduce the amount of energy transm it­
ted to the structure. Floor coverings are most effective in reducing the higher
frequencies of the impact noise. When choosing an appropriate floor finish
many non-acoustic factors have to be considered such as durability and resist­
ance to chemical attack. These requirements can sometimes be m et by install­
ing a floor composed of a hard upper layer and a resilient lower layer e.g. lin ­
oleum on soft fibre board.

Floating Floors

One of the most practical means of obtaining high impact sound insulation
in a building is to use a floating floor construction. A floating floor rests on
the structural floor but is separated from it by a resilient support such as a
mineral wool blanket. The construction can be considered as a mass-spring-
damping element system as shown in Fig.3 .9 . It is vitally im portant in any
floating floor construction that the resilient elem ent is nowhere shorted by a
rigid mechanical connection. Such connections which allow sound to be
transm itted across the element are sometimes referred to as sound bridges.

Fig. 3.9. M echanical analogy of a floating floor structure

The resonant frequency of the floor must be chosen to be very low prefera­
bly less than 20 Hz otherwise, as seen in Fig.3 .1 0 , the insulation at the re­
sonance is less than if the floating floor had not been constructed at all.

75
 Roofing fe lt

fib re

- S kirting board
flo o rb o a rd s 3 cm glass fib re q u ilt

Fig. 3 .1 1 . Examples o f floating floors

76
A low resonant frequency is obtained by having a large floating mass. Above
the resonant frequency the insulation increases at a maximum of
1 2 dB/doubling of frequency when the damping is negligible and at a lower
rate when the damping is increased.

Some examples of floating floors are shown in Fig.3.11.

Particular attention must be paid to electrical conduits, service pipes from

toilets and bathrooms and skirting boards in order to avoid a solid connection
between the structural floor and the floating floor. Pipes and conduits should
be lagged w ith a material such as bitumen fe lt or rubber sheeting where they
pass through the floating flo or and skirting boards should be insulated from
the floating floor by felt or building paper or a sim ilar material.

It should be appreciated th a t a floor construction designed to reduce the

transmission of impact sound to other rooms does not necessarily provide a
low noise level in the room in which the impact source is produced. As an ex­
ample, consider a homogeneous concrete slab structural floor over which is
laid a floating wooden floor. The noise in the source room may be greater for
the wooden floor than for the bare concrete however the noise produced in
other rooms may be reduced because less energy is transm itted to the build­
ing structure.

Ceilings

There are tw o types of ceiling construction which can be used to reduce

the radiation of sound from floors set into vibration due to impacts namely,
the false ceiling and the suspended ceiling. False ceilings are ceilings which
are independent of the main floor ceiling structure. Suspended ceilings are
ceilings w hich are hung from the structural flo o r by wire or resilient hangers.
It should be noted that these ceilings reduce the noise level only in the room
where they are installed and that they do not reduce the radiation of sound
from the side w alls which is due to flanking transmission. They do, however,
improve the insulation both for impact and fo r airborne sounds. Such ceilings
are not usually to be recommended as a means of improving sound insulation
as they are not very effective. Used alone fo r improving the insulation of an
existing floor, the ceiling w ould need to be so heavy that it w ould not be prac­
ticable to construct it. If a floating floor is to be built, nothing extra is gained
by adding a suspended or a false ceiling. If, however, an existing floor cannot
be disturbed and a resilient layer has been laid down to improve the impact
sound insulation then the addition of a heavy suspended ceiling may be used
to give an improvement in the airborne sound insulation.

77
Acoustic Leaks

The expected sound insulation of a w a ll or other building element can be

obtained only if sufficient attention is paid to the insulation of the other parts
of the construction. For example the transmission loss across a w all, de­
signed to give 45 dB of insulation, w ill be fa r below the expected value if the
w all contains ordinary doors and w indow s. Even when doors and w indow s of
special design are used, the measured sound insulation w ill only approach
the expected value providing all the gaps and cracks around the doors and
w indow s are adequately sealed. The higher the sound insulation required,
the more important is th is attention to detail. Less obvious acoustic leaks can
occur via the gaps around service pipes and conduits w here they have to
pass through walls. Transmission of sound via ventilation ducts can also re­
duce the effective insulation of a building element. In buildings w here s u ­
spended ceilings exist it is a mistake from the acoustic point of view for the
dividing w alls between offices to extend only up to the suspended ceiling as
sound can easily propagate from one office to another via the air gap above
the suspended ceiling (see Fig.3 .12a). A sim ilar type of transmission path
can occur when there is a basement or ventilation space below the floor (see
Fig.3 . 12b). These transm ission paths can be blocked by continuing the d ivid ­
ing w all through the air gap to join the main structure.

P Ci? \ I c-. p-k

Suspended Acoustic " le a k "

I
I i 1
Basement or c e l l a r ^ -------------- ----------A coustic " le a k "
I
a)

771147

Fig. 3.12. a) transmission p ath via a suspended ceiling

b) transmission p ath via a basement

Discontinuous Construction

When very high sound insulation is required, a discontinuous construction

is usually employed. This means that the room (or rooms) to be insulated is
completely separated from the main structure of the building supported only
by vibration isolation m ountings which must be designed for the load of that
particular room. It is im portant to ensure that there are no sound bridges or
short circuits by which sound can penetrate into the isolated room. All ser­
vice pipes, ventilation ducts and conduits m ust have flexible joints where
they traverse the cavity. An example of this "box w ithin a box" construction
is shown in Fig.3.1 3.

A irb orn e sound

attenuators
required in
ducting

No rigid
connections via
service pipes
etc. across
cavity

N— V ib ra tio n isolating m ountings

Main structure o f build in g

Fig. 3.13. Discontinuous construction: vertical section through an isolated

room

This type of construction is frequently used for building anechoic rooms

used for acoustic research. For anechoic rooms further acoustical treatm ent
is required inside the room to prevent reflections. This is usually accom­
plished by covering all the surfaces of the room w ith wedges of mineral or
glass wool. The floor can be either a net of steel wire which has a negligible
influence on the free field propagation of sound or a retractable metal grid
which allows one to enter the room to suspend apparatus but the grid is w ith ­
drawn for the duration of the measurements.

79
3.9. VIBRATION CONTRO L

Refrigerators, lifts, air conditioning equipment, heating systems and plum ­

bing are just a few of the many appliances to be found in modern buildings
which can give rise to vibrations and structure borne noise.

If a vibrating machine, pipe or other appliance is rigidly connected to the

structure of a building then structure borne sound waves w ill be transm itted
w ith little attenuation. This structure borne sound can "break o ut" into the
various rooms of the building, producing unwanted noise. This problem is best
solved at the source by using a vibration isolator to reduce the transmission
of energy from the m achine to the structure. If a resilient element is inserted
between the machine and the structure then little transfer of energy w ill oc­
cur providing the isolator is correctly designed. Using ju s t any resilient ele­
ments w ill not suffice. Resilient elements together w ith the mass they sup­
port have resonant frequencies. If these resonant frequencies correspond to
the vibration frequencies of the machine, then energy w ill not only be trans­
mitted but amplified as w ell.

The effectiveness of a vibration isolator is measured in terms of the force

transm issibility (or the displacement transm issibility) w hich is defined as the
ratio of the transm itted force to the applied force. Transm issibility can be ex­
pressed as

1
T - (3.9)

where T = transm issibility

f = driving frequency, Hz
f r = resonant frequency of the isolator,Hz.

The form of this function is shown in Fig.3.1 4.

Most isolators contain damping as w ell as resilience. The effect of the

damping is to reduce the resonance peak but at the expense of reducing the
isolation efficiency at frequencies above the resonant frequency. In many
practical cases, damping is necessary to maintain the amplitude of vibration
w ith in reasonable lim its. This is of particular importance to machines of vari­
able speed which have to be run through the resonant frequency of the sup­
porting isolators.

For the cases where the machine has a particularly annoying resonant fre ­
quency or the machine produces vibrations of an impulsive nature then it can
be advantageous to place the machine on an inertial block (e.g. a slab of con-

80
 forcing frequency
Frequency r a tio = -----------------------------
resonant freq uency 771202

Fig. 3.14. Transmissibility function

crete) which in tu rn is mounted on the vibration isolators. Thus the resonant

frequency is reduced and the inertia of the assembly is increased (Fig.3.15).

The type of isolator used depends on the structure and the environm ent.
Some examples of isolators are steel springs in conjunction w ith oil filled
dash pots, rubber in shear and compression held in a steel foot, cork, felt,
foam and mineral wool mats.

A useful guide for the selecting and applying resilie n t devices is the ISO pu­
blication Draft ISO 2 01 7.

For methods of measuring the vibration due to machines reference should

be made to the ISO publication entitled "M echanical vibration of large rotat­
ing machines w ith speed range from 10 to 200 re v /s Measurement and eval­
uation of vibration severity in situ 3 9 4 5 ", and also to the B & K publication
"Application of B & K equipment to mechanical vibration and shock measure­
m ents".

3 .1 0 . VENTILATION A ND AIR C O N D ITIO N IN G SYSTEMS

Practically every modern building contains some kind of ventilation or air

conditioning system. These systems present a num ber of noise problems
most of which can only be effectively dealt w ith at the design stage. Struc­
ture borne sound is produced by the fan, motor and compressor o f the sys­
tem in the same w ay as from other machines. Aerodynamic noise is pro­
duced by the movement of air in the ducts and through grilles and diffusers.
These noises are transm itted via the ducts to all o th e r parts of the building.

Correct planning at an early staqe w ill save costly modifications later on.
The machinery i.e. motor, fan, compressor, should be placed as fa r as possi­
ble from the areas w hich are liable to be sensitive to noise. The machinery
should be isolated from the main structure of the building by means of anti-vi-
bration mounts (see section Vibration Isolation). S hort lengths of flexible, resi­
lient hosing should be inserted between the m achinery and the ductw ork.

To reduce the noise transmitted by the ducts a num ber of techniques may
be employed such as the lining of the ducts w ith sound absorbing material,
the inclusion of a plenum chamber in the system, provision of bends and
smooth changes of cross sectional area of the ducts, the insertion of ready
made commercially available attenuators and the use of vanes for m aintain­
ing a non-turbulent air flow.

Plenum chambers

These chambers are the only effective remedy fo r low frequency noise. The
chambers should be made as large as possible and should be lined w ith a
thick layer of sound absorbing m aterial. Additional absorption can be ob­
tained by installing baffles w ithin the chamber. The volumes which occur nat-

82
urally in modern buildings e.g. under staircases can be used as plenum cham­
bers.

Lining the ducts

One of the simplest ways of reducing the aerodynamic noise present in the
air flow is to line the ducts w ith sound absorbing material. The linings should
have

1) high absorption coefficient

2) smooth surface for low air friction

3) adequate strength to resist disintegration due to the air stream

4) odourless, fire, rot and verm in proof

It is advantageous to use lining material in the vicinity of a bend in the

duct.

The noise generated by the air flow impinging on a grille (or diffuser) is de­
pendent on the shape of the grille i.e. a badly designed grille w ill induce turb­
ulence in the air flow. If the grille is at fault then it should be replaced w ith
one of better design.

83
 Companies specialising in ventilation systems are aware of the noise prob­
lems involved. It is usually sufficient therefore to specify the acceptable noise
levels in the various rooms of the building and to state the required sound in ­
sulation between the rooms and to leave the design of the ventilation system
up to the suppliers of the ventilation equipment.

84
 4. CRITERIA FOR NOISE CONTROL AND
SOUND INSULATION

4 .1 . INTRODUCTION

In the preceding chapters, the nature and behaviour of sound in rooms and
buildings was discussed. Now these concepts w ill be used to describe the
criteria for the acoustic environment for various human activities. Why
should criteria be necessary at all? The reason can be seen if some of the
more important ways in w hich noise affects people are considered. Noise can

1) damage hearing

2 ) interfere w ith speech communication

3) disturb concentration thus causing a decrease in efficiency

4) annoy.

A t first thought, it would seem that the best environment w ould be one
where there is no noise at all. This is not ti’ue, however, for absolute silence
can be very disturbing and in any case some noise is necessary in offices, for
example, to ensure local privacy. The aim of noise control is to reduce the
noise level in a particular environm ent to an acceptable level and not to rem­
ove the noise altogether.

The acceptable noise level depends on the particular situation. The workers
in an engine room or a foundry or similar noisy work place w ould not nor­
mally be expected to be annoyed by the noise that they make themselves.
There may not even be a need for conversation fo r the work to be done effic­
iently. There is a need, however, to protect the worker against the risk of
hearing damage. In light industry, factories and the like conversation is usu­
ally important. In commercial premises such as offices and shops, the noise
levels present should not interfere w ith conversation and the use of the tele­
phone, nor w ith the concentration of the staff.

The acceptable noise levels in the home are fa r more stringent. Although a

85
particular noise may not cause damage or interfere w ith concentration, it
may be extremely annoying.

4 .2 . HEARING D AM AGE

Noises that are so loud that they cause immediate damage to the ear are
fortunately rare. A gradual deterioration in hearing acuity due to exposure to
excessive noise is far more common. This deterioration in hearing acuity is
usually not apparent to the individual u ntil it is too late and irreparable dam ­
age has occurred. The likelihood of permanent damage occurring is a fu n c ­
tion of length of exposure, noise levels and frequency. This has given rise to
the concept of noise dosage.

The acceptable noise dosage according to the ISO criterion is equivalent to

exposure to a noise of 90dB(A) for an 8 hour working day and a 3 dB(A) in­
crease in the sound level requires half the duration to give the same noise
dose, e.g. 93 dB(A) for 4 hours is equivalent to 90dB(A) for 8 hours w hich is
equivalent to 96dB(A) for 2 hours. The ISO criterion is employed over the
greater part of Europe (Fig.4.1.).

173528

Fig. 4.1. Relationship betw een sound level and duration for the OSHA c rite ­
rion and ISO R 1999 (for ISO applying the criterion 90 dBA for an 8
hour day and no corrections)

86
 In the United States of America the OSHA criterion is employed (Fig.4.1.).
This is again equivalent to exposure to a noise of 90dB(A) for an 8 hour
working day but now a 5dB(A) increase in sound level fo r half the duration
gives the same noise dose, e.g. 95dB (A ) for 4 hours is equivalent to
90 dB(A) for 8 hours w hich is equivalent to 1 00 dB(A) for 2 hours.

The set of Damage Risk Criteria curves (DRC) shown in Fig.4.2. sum m ar­
ises this concept of noise dosage.

Fig.4.2. Suggested Damage risk criteria curves (DRC) based on recommenda­

tion of the U.S. A ir Force 1 9 5 6 (W illiam Burns, NOISE AND MAN,
Murray, 1968)

4.3. N O ISE RATING A N D NOISE C R ITER IA CURVES

W hen talking about noise criteria, it is important to bear in mind that the
recommended sound pressure levels fo r a particular situation are intended as
a guide to the average acceptability of the noise. Human nature, being w hat
it is, w ill ensure that there w ill nearly always be someone who complains no
matter w hat criterion is chosen.

The units most generally used for specifying acceptable noise levels in
buildings are the Noise Criteria developed by Beranek in the United States of

87
 N o t recommended
fo r any office
Large engineering
and d ra ftin g rooms
M ediu m - sized
offices, hotel rooms

Private offices, libraries

Executive offices,
suburban homes
1000 4000
Third-octave freq uency band

Fig. 4.3. Noise criteria curves (NCj

Centre Frequencies o f octave bands Hz

780801

Fig. 4.4. Noise rating curves (NR)

88
America and the Noise Rating adopted by the ISO and used widely in Europe.
The Noise Criteria were developed specifically for commercial buildings and
consist of a fam ily of octave band spectra each w ith its own rating number.
The expected or measured noise in the office or shop is compared w ith the
curves and the lowest curve w hich is nowhere exceeded by the noise gives
the Noise Criteria number, (Fig.4.3). The Noise C riteria curves were originally
intended to relate the spectrum of a noise to the disturbance it causes to ver­
bal communication, including Speech Interference Levels and Loudness Le­
vels. Nowadays, however, the tendency is to specify an acceptable Noise
Criteria number (or a similar criterion) for a particular room instead of talking
about Speech Interference Levels and Loudness Levels.

A similar fam ily of curves has been developed by a working group of the
ISO and is intended for more general application (Fig.4.4 ). Recommended
NR and NC values for various environments are given in tables 4.1 and 4.2
respectively. For further details of noise criteria reference should be made to
the B & K handbook "Application of B & K equipment to Acoustic Noise Meas­
urem ent".

Range of NR Levels
Environment
likely to be acceptable

Workshops 6 0 -7 0
Mechanised offices 5 0-55
Gymnasia, sports halls, swimming baths 4 0 -5 0
Restaurants, bars, cafeterias 35-45
Private offices, libraries, courtrooms 3 0 -4 0
Cinemas, hospitals, churches, small
2 5 -3 5
conference rooms
Class rooms, T.V. studios, large
2 0 -3 0
conference rooms
Concert halls, theatres 2 0 -2 5
Diagnostic clinics, audiometric rooms 10-20
770442

Table 4 .1. Recommended NR values for various environments

89
 Range of NC Levels
Environment
likely to be acceptable

Factories (heavy engineering) 5 5-75

Factories (light engineering) 4 5 -6 5
Kitchens 4 0 -5 0
Swimming baths and sports areas 3 5 -5 0
Department stores and shops 3 5 -4 5
Restaurants, bars, cafeterias and
3 5 -4 5
canteens
Mechanised offices 4 0 -5 0
General offices 3 5 -4 5
Private offices, libraries, courtrooms and
3 0 -3 5
schoolrooms
Homes, bedrooms 2 5 -3 5
Hospital wards and operating theatres 2 5 -3 5
Cinemas 3 0 -3 5
Theatres, assembly halls and churches 2 5 -3 0
Concert and opera halls 2 0 -2 5
Broadcasting and recording studios 15-20
7 /0443

Table 4.2. Recommended NC values for various environments

4 .4 . NOISE IN THE HOME

However low the loudness level of a noise may be, if it intrudes on peo­
ples' privacy in th e ir own homes then the noise is likely to constitute an an­
noyance. NR curves and dB(A) measurements give a good indication of how
much noise people w ill tolerate in th e ir own homes but owing to the highly
subjective character of the problem several corrections have to be applied be-
for a particular NR or dB(A) measurement is acceptable. Various environm en­
tal factors and the nature of the noise itself have to be taken into account.
For example, recognisable audible characteristics of the noise such as pure
tones, hissing, impulsiveness etc, w ill make the noise more annoying than a
broad band noise having the same energy. Further, a given noise w ill be
more annoying to a quiet suburb than in the middle of a big city. The tim e at
w hich a noise occurs (day or night, summer or w inter) also plays a role and
has to be taken into account.

In Britain, recommendations for the acceptability of noise to householders

are given in the British Standard 4 1 4 2 :1 9 6 7 entitled "Method of rating in­
dustrial noise affecting mixed residential and industrial areas". The BS 4 1 4 2

90
is based on the work of the W ilson Committee which tentatively suggested
that the noise levels shown in Table 4 .3 . should not be exceeded in living
rooms and bedrooms for more than 10% of the tim e. Sim ilar recommenda­
tions have been proposed in many other countries.

Environment Day Night

Country areas 40 dB(A) 30 dB(A)

Suburban areas, away from main
45 dB(A) 35 dB(A)
traffic routes
Busy urban areas 50 dB(A) 35 dB(A)
7 70 44 4

Table 4.3. Noise levels in dwellings w hich should not be exceeded for more
than 10% o f the tim e as recommended by British Standard
4 1 4 2 :1 9 6 7

4 .5 . SO UND INSULATION BETWEEN DWELLINGS

To specify the sound insulation betw een dwellings it is not sufficient to use
a single figure index as this could be misleading. The sound insulation is a
function of frequency and should be specified over the frequency range of in ­
terest. It is therefore usual to specify the insulation in the form of a curve

Fig. 4.5. Reference values of sound reduction index for airborne sound be­
tween dwellings according to IS O /R 717 1 96 8

91
 Frequency,Hz 780317

Fig. 4.6. Reference values o f norm alised im pact sound level in octave bands
in dw ellings according to ISO /R 717 1968

and the measured insulation of a building elem ent should not come below
this curve by more than a recommended amount.

The International Standards Organisation has published a Recommendation

entitled "Rating of sound insulation for dw ellings ISO/R 7 1 7 1 9 6 8 " in
w hich the measured airborne sound insulation of a building is compared w ith
the reference curve shown in Fig.4.5., and the measured normalized impact
sound levels are compared w ith Fig.4 .6 .

Examples of national recommendations

Britain

The British system is intended for dealing w ith the dividing w alls (known as
party walls) and floors in blocks of flats and terraced houses etc. The party
w all insulation is based on the insulation of a 9 " brick (0,23 m) w a ll and the
airborne floor insulation is based on the insulation of a concrete floor con­
struction w ith a floating floor fin ish . The measured insulation (corrected to a
reverberation tim e of 0,5 sec w hich is the value of an average living room) is
compared w ith a recommended curve and should fall above th is curve al­
though a total adverse deviation of 23 dB over the 1 6 , one third octave bands
is permitted. The recommended insulation to be provided by party walls in
houses and flats is shown in Fig.4.7.

For impact sound insulation, the British rating system specifies that when

92
Fig. 4.7. Recommended insulation to be provided by party walls in houses and
flats (Great Britain)

Frequency, Hz 771194

Fig. 4.8. Recommended impact sound insulation (Great Britain).

F ull line: bare floor
Dashed line: flo o r covered w ith linoleum

the sound source is a standard tapping machine (see Chapter 5) then the
maximum permitted sound pressure levels in the room below the source
room should not exceed the values shown in Fig.4 .8 . The same conditions
for deviations from the recommended curve apply as for airborne sound insu­
lation.

93
Germ any

The German provisional standard DIN 5221 1 deals w ith both laboratory
and field measurements of sound insulation. The standards used for houses
and flats are shown in Figs.4.9 and 4.1 0.

Frequency, Hz 771193

Fig. 4.9. Recommended airborne sound insulation to be provided between

houses and flats (West Germany)

E
z

</)

Frequency, Hz 7 7119 5

Fig. 4.10. Recommended impact sound insulation betw een flats (West Ger­
many)

94
Denm ark

The Danish M inistry of Housing has published a docum ent entitled "B u ild ­
ing regulation for housing in town and country, 1 9 6 6 " w hich stipulates the
amount of sound insulation required in residential and other buildings. The
measurement of the sound insulation is performed according to the Interna­
tional Standards Organisation publication ISO/R 140 but a correction has to
be applied to make the reverberation tim e of the receiving room a nom inal
0,5 sec.

The document states th a t the party w a lls in blocks of flats should have a
mean insulation of not less than 49 dB and the insulation in any third octave
band should not be less than the values shown in Fig.4.1 1.

Fig. 4 .1 7 . Recommended airborne sound insulation to be provided betw een

dwellings (Denmark)

Standards for w alls w ith in the same fla t, for floors and ceilings between d if­
ferent flats and for w a lls between semi-detached, terraced houses etc. are
also given in Fig.4 .11 .

Other nations

A list (compiled in 1 97 6) of the building standards in 23 countries is given

in the follow ing table.

95
Standards concerning building acoustic C Q Q R Office for S tandards and
from various countries. M easurem ents,
1 134 7 Praha 1,
Argentina Václavské N ám ésti 19
¿SN 3 6 8 8 4 0 M easurem ent of sound insulating
properties of build in g structures
4063 M easurem ent of sound in sulatio n in
d w ellings ÍS N 3 6 884 1 M easurem ent of reverberation time
1974
4065 M easurem ent of sound absorption
coefficients ÍS N 7 3 0 5 2 5 Design in the room acoustics. G eneral
1964 principles
Australia Standards House.
ÍS N 7 3 0 5 2 6 Design in the room acoustics. Studios
8 0 A rth u r Street,
North Sydney 19 6 8 and sound recording rooms
ÍS N 7 3 0 5 2 7 Room acoustics projects. Rooms fo r
A S .1 0 4 5 -1 9 7 1 M easurem ent of absorption 1973 cultural and school purposes. Rooms
coefficients in a reverberation room for public purposes. Adm inistrative
DR 7 5 0 6 0 M ethod of m easurem ent of norm al rooms
incidence source absorption coefficient ÍS N 7 3 053 1 Protection against noise transmission
and specific normal acoustic in building
impedances of acoustic m aterials by
the tube method ÍS N 8 3 0 5 3 5 Sound absorption coefficient
m easurement in reverberation room
DR.7 4 1 6 3 Code of practice for building siting and
construction against aircraft noise Denmark Dansk Standardiseringsråd
in trusio n Aurehøjvej 1 2
DR 7 2 0 9 0 2 9 0 0 H ellerup
Standard method for the rating of
sound insulation for d w ellings D S /IS O R .1 4 0 Felt- og laboratoriem ålinger af luftlyds
og trinlyds udbredelse
DOC.9 7 9 Standard methods for field and
laboratory m easurements of airborne D S /IS O R .3 5 4 Måling af absorptionskoefficienter i
and im pact sound transm ission efterklangsrum
Austria Ö sterreichisches N o rm ungsinstitut Statens Trykningskontor
Leopoldsg. 4
1 0 2 0 W ien Landsbyggeloven, kap. 9, lydisolering
(Building code, sound isolation)
B 81 15 Hochbau, Schallschutz und Hörsamkeit
M iljøstyrelsen
B 8115 Schallschutz und Raum akustik im Kampmannsgade 1
E n tw urf Hochbau 1 604 København V
Ö sterreichischer A rbeitsrin g fü r
Lärm bekäm pfung, P ublikation Støj, bygge-og anlægsvirksomhed
Regierungsgebäude, n r.9 (Noise in build in g construction)
1 0 1 2 W ien P ra n ro L'Association Française de
r r d iio e N orm alisation (AFNOR),
Ö AL-R ichtlinie G eräuscharm e W ohn-, K ranke nhaus­
Nr. 8 und Hotelinstallationen Tour Europe, 9 2 Courbevoie

Ö AL-R ichtlinie S challtechnische G rundlagen fü r die NF S 3 1 - 0 0 2 Mesure, en laboratoire et sur place, de

Nr. 16 E rrichtung von G aststätten und 1956 la transm ission des sons aériens et
Beherbergungsbetrieben des bruits de chocs dans les
constructions
O AL-R ichtlinie Lärm m inderung durch S challschluk-
Nr. 17 kende A usstattung NF S 3 1 - 0 0 3 Mesure du c o e ffic ie n t d'absorption
1951 acoustique en salle réverbérante
Belgium In stitu t Belge de N orm alisation, NF S 3 1 -0 1 1 Code d'essai pou r la détermination en
y 29 av. de Brabançonne,
1974 laboratoire de l'e ffic a c ité des
1 0 4 0 Bruxelles
revêtements de sol en ce qui concerne
N B N 57 6.0 5- M esure en laboratoire de l'indice la réduction des b ru its d'impact
196 3 d 'affaiblissem ent aux sons aériens NF S 3 1 - 0 1 2 Mesure de la durée de réverbération
N B N 57 6.0 6- M esure " in s itu " de l'isole m en t 197 3 des auditorium s
196 3 acoustique aux sons aériens
NF S 3 1 - 0 1 4 Code d'essai pou r la mesure du bruit
N B N 57 6.0 7- M esure en laboratoire de la 19 7 5 émis par les équipem ents hydrauliques
196 4 transm ission acoustique des bruits de des bâtiments
choc
NF S 3 1 - 0 1 5 Mesure du b ru it ém is par la
N B N 576 08- M esure " in s itu ” de la transm ission 197 5 robinetterie de puisage (sanitaire et
1965 acoustique des bruits de choc bâtiment)
N B N 57 6.0 9- M esure du facteur d'absorption S 3 1 -0 1 6 Mesure du b ru it ém is par la
196 8 acoustique en salle réverbérante 1971 robinetterie de bâtim ent
N B N 57 6.4 0- C ritères de l'isolation sonique Germany Beuth Vertrieb GmbH.
196 6 1 0 0 0 Berlin 3 0 , Burggrafenstr. 4 — 7
p r a Q :i Associagäo Brasileira de
(B .R .D . ) und 5 0 0 0 Köln, Kamekestr. 2 — 8
Norm as Técnicas — ABNT DIN 4 1 0 9 Schallschutz im Hochbau
Av. A lm ira n te Barroso 54 Bl. 1 — 5
Rio de Janeiro — RJ
DIN 1 8 041 Hörsamkeit in k le inen bis m ittelgroßen
P M .B .4 3 2 M edida local e em Laboratório de Räumen
Transm issSo de Sons Aéreos e dos
DIN 5 2 2 1 0 Bauakustische Prüfungen.
Ruidos de Impacto (M étodo de Ensaio)
T. 1: Luft- u. Trittschalldäm m ung,
NB 101 Norma para Tratam ento A cú stico em M eßverfahren T. 2: Prüfstände. T. 3:
Recintos Fechados Prüfungen. T. 4: Einzahl-Angaben

96
DIN 5 2 2 1 2 Bestimm ung des Schallabsorptionsgra­ P N -61 Building acoustics. Terminology
des im Hallraum B - 0 2 1 53
DIN 5 2 2 1 4 Bestimm ung der dynam ischen Steifig­ P N -6 8 Building acoustics. Tests on acoustic
keit von D äm m schichten fü r schw im ­ B - 0 2 1 54 properties in building partitions
mende Estriche InspecçSo Gera I dos Produtos
DIN 5 2 2 1 5 Bestimm ung des Schallabsorptionsgra­ Portugal A g ricolas e Industriáis (Repartiçâo
des und der Impedanz in Rohr de Norm alizaçâo)
Avenida de Berna — 1
DIN 5 2 2 1 6 M essung der N achhallzeit in Zuhörer­
Lisboa — 1
räumen
P -6 6 9 Acústica, Ensaio de transmissáo dos
DIN 5 2 2 1 7 Flankenübertragung — Begriffe 1968 ruidos aéreos e de percussâo (airborne
Prüfung des G eräuschverhaltens von and im pact noise transmission)
DIN 5 2 2 1 8
Arm aturen und G eräten der W asserin­ P -6 7 0 A cú stica, Determ inaçâo em cámara
stallation im Labor 1968 reverberante do coeficiente de
absorçâo e da área sonora equivalente
DIN 5 2 2 1 9 Messung von G eräuschen der W asser­
(D eterm ination of sound absorption
installation am Bau
coefficients)
VDI 2 7 1 9 Schalldäm mung von Fenstern
O ficiul de stat pentru Standarde,
G e rm a n y A m t fü r S tandardisiering der D.D.R.,
Roumania Str. Edgar Q uinet 6,
M ohrenstrasse 37a, Bucarest 1
(D.D.R.) 108 Berlin STAS 6 1 5 6 -6 8 Building acoustics. Protection against
TGL 1 0 6 8 7 Bauphysikalische Schutzm aßnahm en, noise and vibration in buildings.
Bl. 1 — 8 Schallschutz. Siehe S .2. Übersicht der Regulation for design and perform ance
Grundlagenstandards STAS 6 1 6 1 -6 0 M ethods of m easurement of noise in
TGL 1 0 6 8 8 M essverfahren der A ku stik buildings
Bl. 1 — 12 STAS 8 0 4 8 -6 7 M easurem ent of dynamic stiffness of
1967 vibration absorbing m aterials in
Great Britain j'p ’áfk l í ^ í ^ lns,ltu,l0n' building acoustics
London W. 1 Sveriges Standardiserings-
B S .2750: Recommendations fo r field and
Sweden kom m ision. Box 3295
1956 laboratory m easurem ent of airborne 1 0 3 6 6 Stockholm
and impact sound transm ission in Bestam ning av ljudisolering i
SIS 025 251
buildings byggnader
BS.CP352: Mechanical ventilation and air Bestam ning av ljudisolering i
SIS 0 2 5 2 5 2
1958 conditioning in b uildings (contains a byggnader, faltm atning
section on sound proofing and an ti­
vibration devices) SIS 0 2 5 2 5 3 Metod fo r vardering av ljudisolering
mellan rum i byggnader
B S .3638: Method for the m easurem ent of sound
1963 absorption coefficients (ISO) in a SIS 8 1 7 3 0 6 Ljudisolerande dbrrar 25 dB, 3 0 dB
reverberation room. och 3 5 dB
BS.CP3: 1 9 7 2 Part 2: Sound in sulatio n and noise Liber Förlag, Fack, 103 2 0 Stockholm
reduction (in buildings)
SBN 75 Ljudklim at
Hungary Magyar Szabványügyi Hivatal, Kap 34
Budapest IX, Üllöl út. 25
Svensk Byggtjänst, Box 1 4 0 3 ,
M E . -8 3 -6 5 Technische V o rschriften des 1 1 1 8 4 Stockholm
M inisterium s fü r Bauwesen
KBS 1 0 -1 9 6 8 Norm er fö r kontorsbyggnader
|t a | Servizio Técnico Centrale,
Y M inistero dei Lavori Pubblici, Switzerland Schw eiz. Ingenieur- und
Roma Architekten-V erein
Circolare N. Criteri di valutazione e collando dei SIA E m pfehlung fü r Schallschutz im W o h ­
176 9 requisiti acustici nelle costruzioni E m pfehlung nungsbau
edilizie 1 8 1 -1 9 7 0
Netherlands Nederlands N orm alisatie-Instituut,

Rijswijk (Z-H)
NEN 1 0 7 0 Sound insulation m easurem ent in U.S.A. A co ustical and Board Products
A sso ciation
197 5 dwellings
2 0 5 W est Touhy Avenue
NEN 2 0 1 4 0 Same as ISO R.1 40-1 9 6 0 Park Ridge, IL 600 68
NEN 2 0 3 5 4 Same as ISO R .3 5 4 A M A - 1-11 M ethod of test. Ceiling sound
1967 tran sm ission test by two-room m ethod
Norway Norges Standardiseringsforbund,
A M Spec A co ustical absorbers
Håkon 7. gt. 2, Oslo 1
No. 1 1 (1972)
NS 3051 * Bestemmelse av lydisolering A ir D iffu sio n Council
NS 4 8 0 4 Måling av lydabsorpsjonsfaktorer i 4 3 5 N orth Michigan
2 .1 9 7 4 klangrom Chicago, IL 60611
P n la n H Polski Komitet Norm alizacji i M iar, A D -6 3 M easurem ent of room -to-room sound
0 ul. Elektoralna 2, (1 9 6 3 ) tran sm ission through plenum air
00-1 39 - W arszawa systems
PN-70 Building acoustics. Soundproof FD -72 Flexible air duct test code
B -0 2 1 51 protection for room s in buildings (1 9 7 2 )

97
 Dept, of Housing and Urban
Development
Am erican National S tandards Institute W ashington, D C.
1 4 3 0 Broadway
A guide to a irbo rne , impact and
N ew York, NY 100 18
structure borne noise control in
S1 7 -1 9 7 0 Sound absorption of acoustical m ultifam ily d w e llin g s
m aterials in reverberation rooms International Conference of Building
Am erican Society for Testing and O fficials
M aterials 5 3 6 0 South W orkm an M ill Road
1 9 1 6 Race Street W hittier, CA 9 0 6 0 1
Philadelphia, PA 1 9 1 0 3 UBC 35-1 Laboratory determ ination of airborne
ASTM Standard method of test fo r impedance sound tran sm ission class (STC)
C 3 8 4 -5 8 and absorption of acoustical m aterials UBC 3 5 -2 Impact sound in sulatio n
(Reapproved by the tube method
1972) UBC 3 5 -3 Airborne sound in sulatio n field test
ASTM Standard method of test fo r sound Komitet Standartov,
C 4 2 3 -6 6 absorption of acoustical m ate rials in
U .S .S .R . Leninsky Prospekt 9 b,
(Reapproved reverberation rooms (ANSI S 1 .7- 117 0 4 9 M oskva M -4 9
1972) 197 0)
Gost 1 5 1 1 6 Sound in sulatio n. M ethod of
ASTM Standard definitions of te rm s relating measurement. S ound insulation factor
C 6 3 4 -7 3 to acoustical tests of building
Gost 1 6 2 9 7 -7 0 Building w ares and materials. The
constructions and m aterials
methods of acoustical tests
ASTM Standard recommended practice for
E 9 0-7 5 laboratory m easurement of airborne Y ugos la via Official Gazette
sound transm ission loss of building
partitions
13 A ug. 1 9 7 0 Regulation on technical precautions
ASTM Standard recommended practice for and conditions fo r sound protection in
E336-71 m easurem ent of airborne sound buildings
in sulation in buildings
ASTM Standard classification for In te r n a tio n a l International O rganization for
Standardization,
E 4 1 3 -7 3 determ ination of sound tran sm ission
class (I. S . O . ) 1, Rue de Varem b£,
Geneva, S w itzerland
ASTM Standard method of testing duct liner
E 4 7 7 -7 3 m aterials and prefabricated silencers R .1 4 0 -1 9 6 0 Field and laboratory measurements of
fo r acoustical and a irflo w perform ance airborne and im pact sound
transmission
ASTM Tenative method of laboratory
E 4 92-73T m easurem ent of impact sound R.3 5 4 -1 9 6 3 Measurement of absorption
transm ission through flo o r c eilin g coefficients in a reverberation room
assem blies using the tapping m achine
(1 9 7 1 ) R.7 1 7 -1 9 6 8 Rating of sound in sulatio n for
dwellings
ASTM Tenative recommended practice for
E 497-73T in stallatio n of fixed p artitions o f light Draft Proposal
fram e type for the purpose of
conserving their sound in s u la tio n IS O /D IS 3 3 8 2 M easurem ent o f reverberation tim e in
efficiency auditories

98
 5. MEASURING TECHNIQUES

5 .1. INTRODUCTION

This chapter describes some of the many applications of Brüel & Kjær in ­
struments to architectural acoustics by describing a generalised chain of
measurement, link by link. The basic theory behind the measurements has
been briefly outlined in the previous chapters but those readers w ho require
a deeper insight should refer to one of the many excellent books on architec­
tural acoustics (see Bibliography). Instrument arrangements w hich can be em­
ployed in order to comply w ith a particular ISO standard, are indicated in the
text (Ref.28).

5.2. GENERALIZED CHAIN OF M EASUREM ENT

Whatever type of acoustical investigation is being performed in a building,

the basic instrumentation required can be considered as a source-transducer-

Fig. 5.1. Schematic diagram o f instrum entation used in architectural acous­

tics

99
analyser chain i.e. one requires a source of sound (or vibration) a measuring
transducer such as a microphone (or an accelerometer) in conjunction w ith a
filte r/a m p lifie r and a recording device. This scheme is shown in a general­
ised block diagram form in Fig.5.1. Each block does not necessarily represent
an individual instrum ent, of course. The whole of the transm itting side of the
diagram, for example, may already be present in the building in the form of a
noisy machine. Furthermore, the entire receiving side may be contained in
one instrum ent such as the Precision Sound Level M eter and Octave A n al­
yser Type 221 5.

5 .3 . SOURCES OF S O U N D AND VIBRATION

For certain measurements, a suitable sound source may already exist in

the room of interest e.g. the organ in a church or the orchestra in a concert
hall may be employed as the sound source for reverberation time measure­
ments. Usually, however, to measure such quantities as sound insulation,
sound absorption and sound power, a suitable sound source has to be intro ­
duced into the room. Three sound sources are produced by B & K for the a ir­
borne excitation of rooms and auditoria. These are the Isotropic Sound
Source Type 4 2 4 1 , the Sound Power Source Type 4 2 0 5 and the Reference
Sound Source Type 4 2 0 4 (fulfills ISO 3741). When measuring the impact
sound insulation of floors, the standardised Tapping M achine Type 3 2 0 4 can
be employed (fulfills ISO 1 40) (Fig.5.2).

The Sound Power Source Type 4 2 0 5 , although developed for sound power
measurements, is an excellent sound source for most purposes in building
acoustics. The 4 2 0 5 is a portable battery operated instrum ent capable of sup­
plying w hite or pink broad band noise. The pink noise can be filtered by any
one of the 7 built-in octave filters whose centre frequencies are 125, 2 50 ,
500, 1000, 2000, 4 0 0 0 and 8 0 0 0 Hz. The sound power output of the 4 2 0 5
can be varied continuously between 4 0 dB and 1 00 dB re 1 pW (1 0 0 d B re
1 pW corresponds to a sound pressure level of 92 dB re 2 x 10“ 5 Pa at 1 m
from the 4205). Among the applications of the 4205 are reverberation meas­
urem ents (as generator can be stopped in less than 30 ms), sound insulation
and sound distribution measurements.

The Reference Sound Source Type 4 2 0 4 is a highly stable, calibrated

source fulfilling the demands of ISO 3 7 4 1 . The 4 2 0 4 is primarily intended
for sound power measurements but is also well suited for sound absorption
and sound insulation measurements.

100
Fig. 5.2. Sources o f noise: Upper left. Tapping M achine Type 3 20 4; upper
right. Isotropic Sound Source Type 4 2 4 1 ; Lower left. Reference
Sound Source Type 4 2 0 4 ; Low er right. S ound Power Source Type
4205

The Isotropic Sound Source Type 4241 is supplied w ith the required signal
from a generator. It is usually necessary to amplify the signal from the gener­

101
ator by employing a power am plifier in order to drive the 4241 or another sui­
table commercially available loudspeaker.

Fig.5.3 shows three signal generators and a Power Am plifier Type 2706.
This power am plifier can be used w ith any of the three generators to give a
power output capacity of 7 5 V A into a 3 Ci load. The Noise Generator Type
1405 can provide the 4241 w ith either a w h ite noise or a pink noise signal.
The w hite noise is used mainly in connection w ith constant bandwidth analy­
sis w hile the pink noise is used mainly in connection with constant percen­
tage bandwidth analysis. An external filte r such as Type 1618 can be used in
connection w ith the generator to produce bands of noise. For frequency re­
sponse and resonant frequency investigations, a sinusoidal signal is required.
This can be supplied by the Sine Generator Type 1023, which can also pro­
duce a warbled tone thus avoiding the establishing of standing waves in a
room.

Fig.5.3. Signal generators and a pow er am plifier : lower left. Noise Genera­
tor Type J 405; upper left. Sine Generator Type 1023; upper right.
Sine Random Generator 1027; low er right, Power A m p lifie r Type
2706

102
 A more sophisticated generator than either the 1405 or the 1 0 2 3 , is the
Sine Random Generator Type 1 0 2 7 . Four types of output signal are available
from this generator. These are w h ite noise, pink noise, narrow band random
noise and a sinusoidal signal. N arrow band random noise is used to avoid the
build-up of standing waves in a room and also w here a swept random noise
is considered to be a more relevant test than a sw ept sinusoidal signal.

All the generators can be rem otely controlled allowing the signal output to
be stopped abruptly as required for reverberation tim e measurements. The
generators all possess compressor circuits which enable the output voltage of
the generators to be controlled by an external voltage. This feature is ex­
trem ely useful in frequency dependent measurements where a given parame­
ter must be kept constant. The compressor can be used, for example, to main­
tain a constant sound pressure level in a room, fo r maintaining the vibration
level of a vibration exciter constant, etc. The generators can be synchronised
w ith the B & K level recorders to produce records of frequency analysis on fre­
quency calibrated paper.

Measurement of quantities such as the loss factor of a wall, the magnitude

of the flanking transmission etc., require that the structure of the building be
excited directly. W hen investigating the effect of impact, it may be adequate
to produce this excitation by a simple hammer b lo w or the fall of a sack full
of sand. Measurement of such quantities as mechanical m obility require a
controllable source of vibration as obtained from a vibration exciter from
B & K's wide range of exciters and auxiliary equipm ent. As exciters vary con­
siderably in their capabilities, it is important to select the exciter having speci­
fications best suited for a particular application i.e. desired frequency range,
sufficient force etc. It is often necessary (e.g. in the investigation of trans­
mission paths) to induce in a w a ll the same level of vibration as th a t induced
by the incidence of airborne sound. These levels o f vibration are usually quite
low so that the vibration exciter employed need not be so pow erful. For this
application, for example a Permanent Magnet Exciter Type 4 8 0 9 may be
used. This exciter has a frequency range of 1 0 to 2 00 00 H z and can deliver
a force of 4 4,5 N peak value and up to 60 N w ith assisted air cooling.

The vibration exciter is driven in the same way as the 4241 described previ­
ously, that is, w ith a signal generator and a power amplifier. The Power Am ­
plifier Type 271 2, which has a power output of 1 8 0 VA, is designed to drive
the Exciter Type 4 8 0 9 safely to its full rating. Other vibration exciter and
power amplifier systems are available which have power handling capacities
less than and greater than the Type 4809.

103
5 .4 . M IC RO PH O N ES AND ACCELEROMETERS

Selection of a Microphone

The sound pressure level produced w ithin the room by the source can be
measured by a calibrated microphone. Several factors have to be considered
w hen selecting the most suitable microphone for a particular application e.g.
the frequency range, the sensitivity required, the directivity and the size.
Brüel & Kjær produce 1 6 different condenser microphones with 4 d ifferent d i­
ameters. The principal features of th is range of microphones are shown in
Table 5.1 together w ith their main areas of application. (Ref. 38).

By choosing the appropriate microphone from this range, precision sound le­
vel measurements can be made in the infrasonic, the audio and th e ultra­
sonic frequency range. B & K microphones have three types of characteristics
i.e. free field, pressure and random incidence. The frequency response of the
microphone above frequencies of about 2 kHz depends on the type o f charac­
teristic the microphone possesses, the differences between the various char­
acteristics becoming more marked as the frequency increases. Standards and
recommendations often specify w hich type of microphone characteristic has
to be employed. Free field microphones have uniform frequency response for
the sound pressure that existed before the microphone was introduced into
the sound field. Any microphone w ill disturb, to some extent, the sound field
in w hich it is placed but the free field microphone is designed to compensate
for its own presence when the microphone is orientated so that the sound ar­
rives perpendicularly to the microphone diaphragm. The pressure micro­
phone, however, is designed to have a uniform frequency response to the ac­
tual sound pressure present which of course includes the microphone's own
disturbing presence. The random incidence microphone is designed to re­
spond uniform ly to signals arriving simultaneously from all angles such as in
the case of highly reverberent or diffuse sound fields. When making measure­
ments in a free field e.g. in the open air or in an anechoic room, a free field
microphone should be pointed directly at the sound source while a pressure
microphone should be orientated at an angle of 9 0 ° to the direction of propa­
gation of the sound so that the sound grazes the fron t of the microphone.

In a diffuse or random sound field, as for example in a fairly reverberant

room, the microphone should be as omnidirectional as possible. In general,
the sm aller a microphone is, the better is its omnidirectional characteristic
and frequency response. However, sm aller microphones are also less sensi­
tive w hich may not be acceptable if the measurements are being made under
relatively quiet conditions. Under these conditions, it is recommended to use
a 1 inch microphone such as the 4 1 4 4 which combines a high sensitivity
w ith a good random incidence response. By mounting specially designed cor­
rectors, such as a random incidence corrector or a nose cone, the response

104
of the 1 , 1 / 2 and 1 /4 inch free field microphones can be made practically in ­
dependent of the angle of incidence over an extended frequency range. Fur­
thermore, the nose cones help to reduce w ind induced noise as experienced
for example, when the microphone is mounted outside the facade of a build­
ing. Another microphone attachment, designed to attenuate noise due to tu rb ­
ulence w hen measuring the airborne noise in airducts, is the Turbulence
Screen UA 0 4 3 6 , which can be fitted to any 1 /2 inch microphone and
preamplifier assembly.

For such applications as acoustic modelling, measuring the sound pressure

in acoustic resonators or near to sound absorbing materials, it can some­
times be advantageous to use the Probe M icrophone Type 4 1 7 0 . The 4 1 7 0
consists of a long narrow probe tube connected to a 1 /2 inch condenser m ic­
rophone and is designed to create the m inim um of disturbance to the sound
field. The Probe Microphone Kit UA 0 04 0, containing probe tubes of various
diameters is designed for use w ith the 1 /2 inch Microphone Type 4134. The
necessary tools and material for cutting and inserting damping material to
make the probe suitable for a particular application are supplied w ith the Kit.

When measuring in diffuse fields, it is im portant to remember that the pres­

ence of the instrum ent case and the operator may obstruct the sound inci­
dent from certain directions and spoil the otherw ise excellent omnidirectional
characteristics of the microphone. Where th is problem is likely to occur the
microphone should be mounted on a flexible extension rod or, better still, on
an extension cable in order to position the microphone as fa r as possible
from the instrum entation and the operator. W hen microphone cables have to
be led through closed doors and windows, a frequent occurence in architectu­
ral acoustics, then the Tape Microphone Cable AR 0001 can be employed.
The thickness of this cable is only 0,2 mm.

In humid or polluted atmospheres, such as in a boiler room, workshop etc.,

it may be preferable to use the M iniature Hydrophone Type 8 1 0 3 instead of
one of the microphones mentioned in Table 5.1.

105
 Frequency
M icrophone Associated S ensitivity
Main area o f applica tio n Diameter range Response
type preamplifier (m V/Pa)
(± 2 d B )

4145 General sound level measurements 2,6 Hz - Free

1" 50
including low sound levels 18 kHz field
4144 2619
Coupler measurements, audiometer 2 ,6 Hz -
1” or Pressure 50
calibration, ca lib ra tion standard 8 kHz
2627

4146 Sonic boom measurements, acoustic 2631 (w ith

pulse measurement, ca rrie r type 1" 2619 same < 0 , 1 Hz -
Pressure 1 2 -6 0
(carrier frequency 10 M Hz) specs, as 8 kHz
4144)

4160 Prim ary calibration standard 1” 2627 2,6 Hz - Pressure 47

8 ,5 kH z
4133 Free field general purpose, general
4 Hz - Free
sound level measurements, electro­ 1 /2 ” 12,5
40 kHz field
acoustic measurements
4134 Pressure, general purpose, general 4 Hz - Pressure
sound level measurements, coupler 1 /2 ” 20 kHz and 12,5
2619
measurements, probe m icrophones random
4149 Best choice fo r p o llu te d o r humid
4 Hz - Free
environments, long te rm o utd oo r 1 /2 ” 12,5
m onitoring systems 40 kHz field

4147 For sonic boom acoustic pulse and 2631 (with Pressure
0,01 Hz -
infrasonic measurements, carrier 1 /2 ” 2619 same and 3 ,7 - 1 8
20 kHz
type (carrier frequency 10 MHz) spec, as random
4134)
4148 Free field general purpose, fo r use 2619
2,6 Hz - Free
w ith low polarizing voltage 1 /2 ” w ith type 12,5
16 Hz field
2804

4165 General free fie ld measurements 2,6 Hz - Free

1/2 ” 7619 50
including low levels 2 0 kHz field
4166 General random sound level 2,6 Hz - Pressure
1 /2 ” 2619
measurements 9 kHz and 50
random
4125 Low cost applications, lo w 5 Hz — Free field
1 /2 ” 2642
polarizing voltage 12,5 kHz and 10
(± 3 d B ) random
4135 Free field sound level measurements 4 Hz - Free
1 /4 ” 4
model work, high levels 2618 or 100 kHz field
4136 2619 +
Random incidence, sound level Pressure
UA 0035 4 Hz -
measurements, boundary layer, pulses, 1 /4 ” and 1,6
70 kH z
coupler, measurements Random
4138 Very high frequency and very high 2618 +
sound level measurements, models, Pressure
UA 0160 6,5 Hz —
1 /8 ” and 1
confined spaces, p o in t source and or 140 kH z
random
receiver, sharp pulses 2619 +
U A 0036

Table 5.1. Principle features and main areas of application o f B & K range o f
microphones

Generally speaking, for almost all sound pressure measurements in b u ild ­

ing acoustics one of the follow ing four h alf inch microphones can be em ­
ployed, the 4 13 3, 4 1 3 4 , 4 1 6 5 and the 4 1 6 6 . For diffuse field measure-

106
merits either the 41 3 4 or the 41 66 can be recommended. Both microphones
have random incidence responses b ut although the 41 66 possesses a grea­
ter sensitivity than the 4134, it has a narrower fla t pressure frequency char­
acteristic. For free field measurements, for example in the direct field of a ma­
chine or in an anechoic room, e ither the 4133 or the 4165 can be recom­
mended. A choice has to be made between the w id e r frequency range of the
4 1 3 3 and the greater sensitivity of the 4165. The 4 1 6 5 is the microphone
that is usually supplied w ith the sound level meters.

Preamplifier Selection

Once the appropriate microphone has been chosen, an accompanying

preamplifier must be selected. The preamplifier presents the microphone with
the correct impedance and also supplies the microphone w ith a suitable pola­
risation voltage. A ll the microphones in Table 5.1 are designed to operate
w ith a DC polarisation voltage of 2 0 0 V except fo r the Type 4 1 2 5 and the
Type 4 1 4 8 which operate with polarisation voltage of 28 V.

The preamplifier has a very high input impedance and presents v irtu a lly no
load to the microphone. A low output impedance enables the connecting
cable between the measuring instrum ent and the pream plifier to be of consid­
erable length. The preamplifiers themselves are powered from e ithe r the
preamplifier input socket of a measuring amplifier or a frequency analyser, or
alternatively from special power supplies if it is required to connect the
preamplifier to other instrum entation.

For 1 inch and 1 /2 inch microphones the Pream plifier Type 2 6 1 9 is recom­
mended. The 1/ 2 inch Type 4 1 4 8 can be used w ith the Type 261 9 providing
it is powered from the two channel, battery operated Microphone Power Sup­
ply Type 2804.

The 1 /2 inch Type 41 25 has been especially designed for low cost applica­
tions. It is used w ith the Pream plifier Type 264 2 and the two channel, bat­
tery operated Microphone Power Supply Type 2 81 0.

The 1 /4 inch and the 1 /8 inch microphones are designed to operate in con­
junction with the Preamplifier Type 2 61 8. The figure overleaf shows some of
the microphone equipment already described.

107
Fig. 5.4. Microphones and pream plifiers: top left, 1 / 8 " M icrophone + UA
0 16 0 + Pream plifier Type 2618, top centre, 1 /2 " M icrophone +
Pream plifier Type 2 6 1 9 ; top right, 1 " Microphone + A daptor DB
0 3 7 5 + Preamplifier Type 2619; lower. Probe M icrophone Type
41 70 w ith built-in pream plifier

For powering the preamplifier, one of the three power supplies show n be­
low can be employed.

F ig.5.5. Microphone Power Supplies; left to right. Types 2801, 2 8 0 4 and

2807

108
The mains driven Power Supply Type 2801 can supply all voltages for the
microphone assemblies using 2 0 0 V polarisation voltage. The battery driven
Power Supply Type 2 80 4 can be used only w ith the Preamplifier Type 2619.
It supplies all the necessary voltages for two microphone assemblies and can
be adjusted to give 28 or 2 0 0 V polarisation voltage. The mains operated,
tw o channel Power Supply Type 2 8 0 7 can supply all voltages for tw o micro­
phone assemblies using 2 00 V polarisation voltage and allows automatic
switching between the measuring points, a very useful feature w hen measur­
ing, for example, the sound insulation between tw o rooms.

M icrophone 412 5 M icrophone 4134

/ \
M icrophone ■P ream plifier 2619
— Pream plifier
2642
7~T" to measuring a m p lifie r
M icrophone Power
Supply
2810
LJf T w o Channel M icrophon e
Power S u pply
2807

T w o Channel
Level
Recorder
2309

Fig. 5.6. The use o f two microphones w ith a single microphone pow er supply

Calibration of Microphones and Associated Measuring Systems

Microphones and their associated measuring systems can be rapidly and ac­
curately calibrated by using one of the two follow ing instruments either the
Sound Level Calibrator Type 4 2 3 0 or the Pistonphone Type 4 2 2 0 . These are
battery driven instruments w hich produce a known sound pressure level at a
known frequency. The accuracies obtainable from calibrating w ith these in­
struments are ± 0 ,2 5 d B and ± 0 ,2 d B respectively. The M iniature Hydro­
phone Type 8 1 0 3 can be calibrated as easily as the microphones by employ­
ing the Hydrophone Calibrator Type 4223.

It is good policy to calibrate the instrum entation before and after each set
of measurements. If a tape recorder is employed, a calibration tone can be re­
corded on the magnetic tape together w ith a verbal description of the ampli­
fie r settings.

109
Fig. 5.7. Upper, Sound Level Calibrator Type 4 2 3 0 ; lower, Pistonphone Type
4220

Selection of an Accelerometer

The vibration of w alls, windows, pipes and other elements in a building

can be measured by attaching an accelerometer to the vibrating structure. An
accelerometer is an electromechanical transducer w hich gives, at its output
term inals, a voltage proportional to the acceleration to which it is subjected.
(Ref. 39).

The choice of an accelerometer depends on such factors as the type of v i­

bration to be measured, the level of vibration, the frequency of the vibration
and the mass of the vibrating structure. In order not to change the character
of the motion of the vibrating structure, the accelerometer must have a mass
w hich is small compared to the mass of the vibrating structure. In architectu­
ral acoustics, this is usually the case as one is dealing w ith massive struc­
tures such as w alls and floors. W indow s can, however, possess a very low
mass and therefore th eir vibration characteristics can be influenced considera­
bly by the addition of a mass in the form of an accelerometer. In these cases
a very lightweight accelerometer should be used. W hen mounting acceler­

110
ometers on w alls, floors etc., the simplest method is to use bees' wax. A thin
layer of wax is spread on the surface and the accelerometer pressed firm ly
onto the wax. The great advantage w ith this method is that the accelerome­
ters can easily be removed and rapidly placed elsewhere. A nother simple
method is to attach the accelerometer to the structure w ith double sided
adhesive tape. Other methods of mounting include the use of epoxy based ce­
ments and the use of accelerometer w ith magnetic bases for m ounting on me­
tal panels etc. Every B & K accelerometer is individually calibrated and is deliv­
ered w ith an individual calibration chart. The calibration of an accelerometer
and the complete measuring system can however be checked by using the
Calibrator Type 429 1.

In building acoustics, it is often the low frequency vibrations of w alls and

other massive structures that are of interest. It can therefore be useful to li­
m it the high frequency response of the accelerometer by inserting the Me-

Fig. 5.8. A selection o f B & K accelerometers

1 11
chanical Filter for Accelerometers Type U A 0 5 5 9 between the accelerometer
and the mounting point. The cutoff frequency depends on the accelerometer
mass but the — 3 d B point is typically 4 kHz w ith an accelerometer Type
4 3 7 0 w hich weighs 54 g. Additional masses can be added to regulate the cut-
of frequency. Fig.5 .8 shows a selection of B & K accelerometers.

The principle features of the B & K range of accelerometers are show n in

Table 5 .2 together w ith their main areas of application.

Charge Voltage Freq. Range M ounted W eight

T yp e Sensitivity S e ns itiv ity to + 10% lim it Resonance Main area o f a p p lic a tio n
No. (p C /m s -2 )* ( m V /m s -2 )* (Hz) A (kHz) (grams)

436 6 - 4 ,5 ~ 4 0 ,2 - 700 0 22 28

4367 ~ 2 ~ 1,5 0,2 - 970 0 32 13

General purpose
436 8 ~ 4,5 ~ 4 0 ,2 - 700 0 22 30

436 9 ~ 2 ~ 1,5 0 ,2 - 970 0 32 14

4370* 10 ± 2% - 10 0 ,2 - 60 0 0 18 54
Low level measurements
4371* 1 ± 2% ~ 1 0 , 2 - 12000 35 11
M iniature accelerometers
43 4 4 ~ 0,25 ~ 0,25 1 -2 1 0 0 0 70 2 fo r high level and high
frequency measurements.
0,4 Ideal fo r delicate s tru c ­
830 7 ~ 0,07 ~ 0,22 1 - 250 00 75
excl. cable tures and c o n fin e d spaces
Triaxial. For measurements
4321* 1 ± 2% ~ 0,8 1 - 12000 40 55 in three m u tu a lly perpendi­
cular directions.
Standard reference
8305 ~ 0,12 - 0 - 4 4 0 0 (2 % ) 30 40 accelerometer.
Low frequency and lo w
8306* 10 0 0 ± 2%* 1000 ± 2 % 0 ,2 - 1000 4,5 500
level measurements.
100 Permanent v ib ra tio n
83 0 8 » 1 ± 2% ~ 1 1 -1 0 0 0 0 30
m onitoring on in d u s tria l
8310* 1 ± 2% ~ 1 1 - 10000 30 100 excl.cable machinery etc.

8309 - 0 ,0 0 4 ~ 0,03 1 - 60000 180 3 excl. cable High level shocks.

• Uni G ain® types

* 1 m s- 2 = 0,1 g
A Low er lim itin g frequency d ete rm in e d by preamplifier used and environmental conditions

Table 5.2. Principle features and m ain areas o f application of the B & K
range o f accelerometers

Preamplifier selection

Accelerometers, just like microphones, have to be followed in the m easure­

m ent chain by a preamplifier. The pream plifier is necessary for tw o reasons.
Firstly, to convert the high output impedance of the accelerometer to a lower
value suitable fo r input to measuring amplifiers and analysers; secondly, to
am plify the relatively weak output signal from the accelerometer if th e follow ­
ing instrum entation does not have a sufficiently high sensitivity.

112
 There are basically tw o kinds of am plifier, namely the voltage amplifier and
the charge amplifier. The voltage am plifier is designed to present the highest
possible resistance to the accelerometer w hile the input capacitance is kept
low. The voltage am plifier is usually placed near the accelerometer w ith only
a short interconnecting cable in order to avoid loss of sensitivity due to the
capacitance of the cable. The voltage am plifier allows the use of long cables
between the preamplifier and the measuring amplifier. The charge am plifier
is designed to present both a very high input capacitance and resistance to
the accelerometer. W ith the charge am plifier, the variation in input signal
due to the varying capacitative loading of the accelerometer is compensated
by a capacitance feedback from the pream plifier. An advantage of the charge
am plifier therefore, is that very long cables can be used between the acceler­
ometer and the pream plifier w ithout changing the sensitivity of the m easur­
ing system. Nowadays, the most widely used accelerometer preamplifier is a
charge amplifier. The table below shows the main features of the voltage and
charge amplifiers in the B & K range.

It is not necessary for the accelerometer to be followed by a preamplifier in

the follow ing cases: w hen using the high sensitivity accelerometer Type
830 6 w hich has a built in preamplifier and when the signal from the acceler­
ometer is fed into measuring instrum ents possessing a b uilt in pream plifier
such as the vibration meters and the sound level meters.

V olta g e A d j. Power F requency Meas. F ilt ers

T yp e o r Charge Gain S u pp ly Range M ode LP HP F eatures A p p lic a tio n

AC 0,3 Hz 3 D ig it sensitivity conditioning Multi-purpose, vibration

2626 Charge X mains to Acc. X X W ide range o f output levels. m onitoring and measurement
100 kHz T w o signal level indicator lamps.
D irect o r transformer o utp ut

Ext. 1 Hz Small, rugged construction Multi-channel, vibration

2634 Charge X Batt. to Acc. - X Suitable fo r single ended and m onitoring and measurement
200 kHz d iffe re n tia l output transducers
Ext. 0,1 Hz Acc. 3 D ig it sensitivity Multi-purpose, vibration
2635 Charge X or to Vel. X X adjustment. m onitoring and measurement
Int. 100 kHz Dis. Wide range of output levels.
Batt. B u ilt-in test oscillator

Voltage AC 0,3 Hz 4 D ig it sensitivity adjustment Calibration and reference w o rk

2650 & X mains to Acc. X X Wide range of output levels
Charge 200 kHz T w o signal level indicator lamps
B u ilt-in test oscillator

Ext. 0,003 Hz Acc. G rounded or Floating input. Multi-channel, vibration

2651 Charge X Batt. to Vel. - X O verload indicator. m onitoring, vibration severity
200 kHz measurement ,ao,4s

Table 5.3. Principle features of the B & K range o f accelerometer pream plifi­
ers

1 13
Calibration of a vibration measuring system

The accelerometer and its follow ing chain of measuring instrum ents can be
simply calibrated by employing the Accelerometer Calibrator Type 4291. This
calibrator is a battery operated instrument w ith a built in shaker table produ­
cing a reference acceleration level of 1 0 m .s .—2 peak at a frequency of
79,6 Hz ( = 5 0 0 rad/s). The calibration method involves vibrating the accel­
erometer at 10 m .s.~2 peak and adjusting the sensitivity of the indicating in ­
strum ent for a fu ll scale peak meter indication of 1,0 or for an rms indication
of 0 ,7 0 7 . The measuring sytem w ill then give a direct indication of the level
of acceleration in m .s —2 peak or m.s.—2 rms. (Ref. 40).

Fig.5.9. Accelerom eter calibration : left to right. Accelerom eter Calibrator

Type 429 1 with accelerometer m ounted on the vibration table; Condi­
tioning A m plifier Type 2 6 2 6 ; M easuring Am plifier Type 2 6 0 7

5.5 . AMPLIFIERS, FILTERS, ANALYSERS A N D RECORDERS

Portable instruments for sound measurement

For field measurements, portable instrumentation is an advantage. In many

situations a sound level meter can be used in conjunction w ith a portable fil­
ter set thus combining the microphone, am plifier and filte r in a single easily
manageable instrument.

1 14
 All the instrum ents shown in Fig.5.10, except the 2 21 9, have an output
for use w ith chart or tape recorders. The 2 2 1 0 , 2218, 2 2 0 9 and the 2 2 0 3
can be used in conjunction w ith filter sets 1613 and 1 61 6. It is often the
case that a straight forward measurement of the sound pressure level w ithin
a room, is not sufficient and that a time average of the sound pressure level
is required. This time average can be determ ined by using either the Preci­
sion Integrating Sound Level Meter Type 2 2 1 8 or the Noise Level Analyzer
Type 4 42 6.

N.B. If the 4 4 2 6 be used to determine the mean sound pressure level, L,

then the microphone must be powered from a source other than the 4 4 2 6 .
The microphone output is th en fed into the 4 4 2 6 via the "D ire c t" BNC socket
on the rear panel of the instrum ent. This is necessary because the "Pream pli­
fie r” input of the 4426 is A-weighted so th a t if this input w ere used, the re­
sult would be Leq and not L.

Precision w ith O utput fo r R ecorders

-\ General
Filter C onnection purpose
________ A _________
Impulse
____A____

Q!» §

2210 2218 220 9 220 3 2206 2219

(Integrating)

Fig. 5.1 O.a. The B & K range o f Sound Level M eters

® •• ••
•- —
& • *• • V
• • • .=
1613 1616 1621 1623
Octave 1/3 O ctave 3 and 23% 5, 10 and 23%
F ilte r Set F ilte r Set T u n a b le T ra c k in g

Fig. 5 . 10.b. Portable filters

1 15
Fig. 5 .1 7. Time averaging o f sound pressure: left, Integrating Sound Level M e ­
ter Type 2 2 1 8 fitte d with Octave Filter Set Type 1613; right. Noise
Level Analyzer Type 4426

Portable instruments for vibration measurement

An accelerometer can be used in conjunction w ith a sound level meter

(types 2 2 0 3 , 2209, 2 2 1 0 or 221 8) and an Integrator ZR 0 0 2 0 to yield a por­
table system for measuring the acceleration, velocity and displacement due to
vibrations. To analyse the vibration signal in frequency bands either the Third
Octave or Octave Filter Type 1616 or 1 6 1 3 can be directly attached to the
bottom of the sound level meter (Fig.5.1 2).

For more detailed on site analysis, the Portable Vibration Analyzer Type
3513 is recommended. This self contained system consists of the Vibration

116
Fig. 5. 12. Accelerom eter used in conjunction w ith a sound level meter, a fil­
ter set and an Integrator Type ZR 0 0 2 0 to yield a portable system
for analysing vibrations in th ird octave bands

Meter Type 251 1 and the Tunable Band Pass Filter Type 1621 mounted in a
hard plastic carrying case perm itting analysis in 3% and 23% (i.e. third oc­
tave) frequency bandwidths.

Fig. 5.13. Portable Vibration Analyzer Type 3513

117
Measuring amplifiers and filters

In the laboratory, the signal from the transducer (microphone or acceler­

ometer) can be measured and analysed by mains operated instrumentation
such as a combination of a Measuring Am plifier Type 2 6 0 6 , 2607 or 2 60 8
and a Band Pass Filter Set Type 1 6 1 7 or 1618. These three am plifiers are
basically low noise, w ide range, calibrated voltmeters and have many com­
mon components, the main difference being in their rectifying circuits. A ny of
these amplifiers may be employed w ith either of the tw o filter sets. The var­
ious combinations give different measurement and analysis possibilities. The
basic system, for example consisting of the Type 1 6 1 8 and the Type 2 6 0 8 is
intended for rms measurements in the audio frequency range w ith the option
of filte r selection being performed by a level recorder. The optimum use of all
the measuring modes and control possibilities is obtained with a combination
of the Type 1617 and Type 2607 enabling measurements and analysis to be
performed in the frequency range from 2 Hz to 1 6 0 kHz obtaining rms and
peak values from all commonly encountered random, quasi-random, periodic
and complex waveforms.

Frequency Analyzers

In the generalised chain of instrum entation shown in Fig.5.1, the filte r and
am plifier on the receiving side are often contained in a single instrum ent
known as an analyser.

For analysing a signal w ith a tunable constant percentage band pass filter
of bandwidth 1%, 3%, 10% and 23% (i.e. one third octave) the Frequency A n ­
alyzer Type 2 1 2 0 and Type 2121 are recommended. For narrow band fre­
quency sweeps, for example, in determ ining the frequency response of build­
ing elements to an applied force or the modal response of a room to a swept
pure tone, the Heterodyne Analyzer Type 2010, consisting of a frequency se­
lective measuring am plifier and a beat frequency oscillator, is recommended.

A much more sophisticated instrum ent for the analysis of signals is the
Digital Frequency Analyzer Type 2 1 3 1 . This instrum ent enables signals to be
analysed in real tim e, in octave or in third octave bands and the spectra can
be displayed on a screen in the form of a bar graph. The spectra are then
stored for later recall and comparison. W ith this facility, for example, the
spectra of the noise on either side of a wall in sound insulation measure­
ments, can be displayed simultaneously on the screen.

On coupling the 2131 to a programmeable desk top calculator, the possible

applications of the instrum ent are greatly increased. Measurements of sound
insulation and reverberation time fo r example can be fu lly automated, the re­
verberation decay curves being displayed on the screen.

118
 For analysing either sound or vibration signals in narrow band and for the
capturing of transient signals, the N arrow Band Spectrum Analyzer Type
2031 is recommended. The 2031 has a screen where either the tim e h is­
tory, the instantaneous spectrum or the averaged spectrum can be displayed.

Fig.5.14. Frequency analysers: upper left. Heterodyne Analyzer Type 2 0 1 0 ;

low er left. Frequency Analyzer Type 2120; upper right. D igital Fre­
quency Analyzer Type 2131; lo w e r right. M easuring Am plifier Type
2 6 0 8 in combination with the F ilte r Set Type 161 8

Tape Recorders

It is often convenient to record the signals from the transducer on magnetic

tape and to perform the analysis later in the laboratory. Two tape recorders
are produced by B & K w hich can be used for this purpose, the Type 7 0 0 3

119
and the Type 7 00 4. The 7 0 0 3 is a four channel frequency modulated recor­
der intended principally for vibration measurements and has tw o speeds for
frequency transform ation, 1,5 ips and 1 5 ip s . Using the 1 5 ips, the recorder
has a linear frequency response up to 1 2,5 kHz which is sufficient for most
measurements in building acoustics.

Chart Recorders

During the analysis of the signal, a permanent recording of the results can
be obtained by synchronising a chart recorder w ith the analyser or the gener­
ator in order to produce a frequency scan, a decay curve, a tim e function etc.
The chart recorder can be a level recorder such as the Types 2 30 6, 2307 or
the 2 3 0 9 or the X-Y Recorder Type 2 30 8. The Level Recorders Types 2 30 6
and 2 3 0 9 are battery operated and possess one and two channels respec­
tively. A ll three level recorders are capable of recording on both linear and lo­
garithmic scales.

Fig.5.15. Chart Recorders: top left. Level Recorder Type 2 3 0 7 ; top right, X-Y
Recorder Type 2 3 0 8 ; lower left. Two Channel Level Recorder Type
2 3 0 9 ; lower right. Level Recorder Type 230 6

120
 The X-Y Recorder Type 2 3 0 8 is mains operated and is intended for the lin ­
ear recording of DC signals on paper of A 4 size. When the analysis is per­
formed by an instrument such as the Statistical Analyzer Type 4 4 2 6 or the
Digital Frequency Analyzer Type 2131, the results can be printed out by the
Alphanumeric Printer Type 2 31 2.

Calculators

Many acoustical measurements which w ould otherwise be repetitive and

longwinded can be greatly simplified by employing a Digital Frequency Anal­
yzer Type 2131 in conjunction w ith a programmable desk top calculator. The
calculators recommended by B & K for use w ith the 2131 are the Tektronix
4051 and the Hewlett-Packard HP 9 8 2 5 A (R ef.30). Two program packages
are produced by B & K for performing basic acoustic measurements such as
reverberation tim e, sound power, Leq etc., w ith these calculators. One pro­
gram, the BZ 001 1, is for use w ith the 21 3 1 /4 0 5 1 system and the other BZ
001 2 is for use w ith the 2 1 3 1 /9 8 2 5 A system. A listing of the programs
available on BZ 001 1 is shown in Fig.5.1 6 . The Narrow Band Spectrum Anal­
yzer Type 2031 can also be used in conjunction w ith a calculator.

B & K

B & K A C O U S T IC A L C A L C U L A T IO N S

PR O G RAM PACKAG E
BZ 0011

Briiel & Kjæ r

Place the tem p late fo r the USER D E F IN A B L E KEYS

and make y o u r choice on these keys:
KE Y 1: This program.
2: In p u t a selectable num ber o f spectra fro m the 2131
and calculate the average.
3: D ifference between the averaged spectra in A & B.
4: Calculate PNDB o f each spectrum.
5: Stevens calculation on the averaged spectrum.
6: 3-D p lo t o f the num ber o f spectra in A o r B buffe r.
7: Reverberation tim e.
8: Sound power. - Precision m ethod fo r broad band
sources in reverberation rooms (ISO 3741).
9: L calculation.
10: 1/12 octave in p u t fro m 2131.
PLEASE, M A K E Y O U R CHOICE!

Fig. 5.16. Listing of the programs available on the Program Package BZ 0011

121
 6. SUGGESTED INSTRUMENTATION

6 .1 . INTR O D UC TIO N

The foregoing chapter describes the various instrum ents in a generalised

measurement chain. Now various instrum ent arrangements w ill be described
w hich are intended for particular architectural acoustic measurements. It
should be borne in mind, however, th a t the instrum ent arrangements and the
measurements which are mentioned in th is section constitute by no means
an exhaustive survey of all the possible applications of B & K instrum ents in
th is field.

6 .2 . REVERBERATION TIME

One of the most important characteristics of a room or auditorium is its re­

verberation time. To a large extent it is the reverberation time which deter­
mines the suitability of a room for a particular purpose. Reverberation cham ­
bers are specially designed rooms found in most acoustic laboratories. Such
chambers have hard, non-parallel surfaces in order to have the m inim um
sound absorption and therefore the longest reverberation time possible, often
in excess of 12 seconds. Reverberation chambers are used for the m easure­
m ent of the absorption coefficients of materials and for the determ ination of
the sound power of noise sources. Long reverberation tim es are also encoun­
tered in stairwells and in churches. Another special type of room used for
measurements of directivity and the determ ination of sound power is the ane-
choic chamber. All surfaces of such a chamber are covered w ith sound ab­
sorbing material thus preventing the reflection of sound energy w hich yields,
ideally, an enclosure where the reverberation tim e over the whole audio
range is zero. The reverberation tim es usually met w ith in architectural acous­
tics fall between these tw o extremes (see Fig.2.1 2).

The technique for measuring reverberation time depends on the particular

circumstances but the procedure is basically the same, i.e. the sound source
is introduced into the room the source is abruptly stopped and the m icro­
phone monitors the decay of the sound pressure level w ithin the room. The
decay in sound level is recorded by, fo r example, a level recorder and the re­

122
verberation time is measured from the slope of the decay curve. This proce­
dure can be repeated for each frequency band of interest.

It is sometimes possible to utilise those sound sources which already exist

in the room,e.g. the organ in the case of a church, the orchestra in the case
of a concert hall. However, the room is usually excited by one of the fo llo w ­
ing methods

1) Pistol shot
2) Wide band noise
3) Narrow band noise

Each method has its advantages and disadvantages. Probably the simplest
way of exciting a room is to use a pistol shot. This method has, however,
gone out of favour because of its lack of reproducibility and its lack of energy
at the lower frequencies.

A better method of excitation is to use electronically generated noise con­

sisting of a wide band of frequencies delivered to the room via an am plifier
and a loudspeaker. Two types of w ide band noise are in common use; w hite
noise which has constant energy per unit of frequency and pink noise whose
energy content is inversely proportional to frequency (i.e. — 3dB per octave
or — 10dB per decade). It is usual to filte r wide band noise into octave or
third octave bands before delivering the signal to the loudspeaker in order
that a greater sound pressure level can be obtained in the frequency band of
interest for a given loudspeaker w ith a given power handling capacity. For ex­
ample, consider a loudspeaker w ith a fla t frequency response over eight oc­
taves. Assume that the maximum sound pressure level obtainable in a parti­
cular octave band is x dB when the loudspeaker is fed w ith wide band pink
noise. When this noise is restricted to the octave band of interest by the use
of filters then as the bandwidth is reduced from 8 to 1 octave (i.e. 2 x 2 * 2)
then the maximum sound pressure level obtainable is increased by 9 d B to
(*+ 9) dB that is, by 3 dB for each halving of bandwidth.

A band of noise is far more satisfactory than pure tone excitation because
a pure tone may w e ll overemphasize the natural resonances of the room. A
band of noise is necessary in reverberation measurements in order to obtain
an average of the contributions of all resonances of the room to the total
sound pressure level in a particular band of frequencies. Another way of over­
coming the problem of room resonances is to use the warbled tone fa cility of
the B & K generator Type 1023. W hatever sound source is employed, the
sound pressure levels produced in the room should be at least 4 0 dB above
the background noise level in all the frequency bands of interest in order to
obtain an adequate decay curve.

When measuring the reverberation tim e of a room in a laboratory then one

123
may, w ithin reason, take as long a time as necessary and the instruments
may remain in the room from one day to the next. Thus making measure­
ments in 1 5 third octave bands for several positions of the microphone in the
room poses no special problem. However, for an occupied living room or in a
newly constructed fla t on a building site, the measurements have to be per­
formed as quickly as possible. For on site measurements, portable equipment
is highly desirable as a suitable mains supply of electricity is not always avail­
able. One way of saving valuable tim e is to record the on site measurements
on a tape recorder and perform the necessary analysis in the laboratory.

Problems arise w hen measurements have to be made in the presence of an

audience. One cannot reasonably expect an audience who have paid to listen
to a concert, for example, to endure a measuring technique w hich employs
bursts of noise in 1 5 frequency bands. In such cases a single burst of pink
noise or a pistol shot would be suitable alternatives, the sound pressure level
decays being recorded on a tape recorder for analysis later.

Having chosen a sound source, one then has to decide where to position it.
In some rooms the choice of the position of the source is obvious, e.g. in the
pulpit in the case of a church, at the lectern in a lecture theatre and on the
stage in a concert hall. Other positions of the source around the room help to
locate excessively long RTs due to room modes.

In small rectangular rooms, the best position for the sound source is in a
corner of the room, i.e. where all the natural frequencies of the room pos­
sess a pressure maximum.

In small rooms a single value of reverberation tim e is sufficient to charac­

terise the reverberation of the room but in large rooms and auditoria this is
not possible as the reverberation tim e can vary considerably from place to
place throughout the volume of the room. In a thorough investigation of the
acoustical properties of such rooms therefore, the reverberation decay should
be monitored at several positions over the area where the audience are
seated.

The following four instruments arrangements all comply w ith ISO 3382
and ISO/R 354.

Portable Arrangement — Pistol Shot Method

This method has the advantages that it is quick and the necessary instru­
mentation is kept to a minimum (Fig .6.1). The disadvantages are that the
noise spectrum is limited in frequency content and lacks reproducibility. To
check that sufficiently high sound pressure levels are produced at both high
and low frequencies, it is advisable to frequency analyse the impulse before

124
 Fig. 6.1. Portable arrangement - Pistol shot method

making any reverberation measurements. This analysis should preferably be

performed in an anechoic room.

Portable Arrangement — Filtered Noise Method

The instrumentation in Fig.6 .2 can be used to measure the reverberation

tim e in the seven octave bands from 125 Hz to 8 0 0 0 Hz and in wide band. If
the reverberation tim e needs to be determined in th ird octave bands then the
Sound Level Meter Type 2215 can be replaced by a Sound Level M eter Type
2 2 0 3 and a Third Octave Filter Set Type 1616. The w riting speed of the
2 3 0 6 , limits the shortest reverberation time measurable w ith this arrange­
ment to 0 ,24 s. For shorter reverberation times either the Level Recorder
Type 2307 should be used which is however mains driven, or the reverbera­
tion decays should be recorded on tape and analysed later in the laboratory.
For measuring the reverberation in several positions in the auditorium , it is of­
ten convenient to record the reverberation decays on a tape recorder and to
analyse the recordings later.

Fig. 6.2. Portable arrangement - Filtered wide band noise m ethod

125
Automatic Arrangement - Paper Loop Method

When many reverberation tim e measurements are required at various posi­

tions in the auditorium and the portability of the instruments is not important
e.g. in concert halls, theatres, churches etc., it is advantageous to use an au­
tomated system such as the one shown in Fig.6 .3.

Fig. 6.3. Automatic arrangement - Paper loop method

1 26
The generating side of the arrangem ent consists of a random noise generator
w hich sends a w h ite noise signal through the filte r and am plifier to the
Sound Source Type 4 2 4 1 . The receiving side consists of a microphone, am pli­
fier, filte r set and a Level Recorder Type 2307. The filte rs on both generating
and receiving sides are set m anually to the same bandwidths and centre fre­
quencies. The level recorder then automatically operates in the correct se­
quence the starting and stopping of the generator, the stepping of the filters
in unison and the lifting and low ering of the recorder pen. When the sound
source is switched off, a decay curve is traced by th e level recorder. The re­
verberation time is determined from the slope of the trace which represents
the decay in dB per second. A special protractor, SC 2 36 1, is supplied for
this purpose. To obtain a series of curves similar to those shown in Fig.6 .3,
the recording paper should be made into a loop as shown in Fig.6 .4 consist­
ing of two chart lengths less one sprocket hole. The start of each decay curve
is separated from the start of the next by 5 mm w hich corresponds w ith the
third octave spacings on the preprinted paper. If th ird octave filtering be used
then each decay curve is representative of a different third octave band. If oc­
tave filtering be used then for each octave band there w ill be three sim ilar de­
cay curves.

Fig. 6.4. Autom atic arrangement - Loop of recording paper is fitted on the Le­
vel Recorder

127
Automatic Arrangement — Digital Frequency Analyser/Calculator Method

This method employs a Digital Frequency Analyser Type 2131 and a Noise
Generator Type 1 40 5 which are controlled via an IEC interface by a desk top
calculator containing a B & K Acoustic Program Package Type BZ 0011 or
001 2. Once the controls of the 1 4 0 5 and its associated amplifiers have been
set manually, the system may be programmed to automatically repeat a cycle
of operation a predetermined num ber of times. The cycle of operation con­
sists of the noise source being switched on for a long enough tim e for a
steady sound field to be established in the room; the source is switched off
and the spectrum of the decay is monitored by the 2131 and transm itted to
the calculator; the noise source is then switched on again and the cycle is re­
peated.

During the decay the 2131 transm its spectra to the calculator at selected
read-in intervals, the shortest interval being 4 4 ms. As the read-in process
can always be referred to a definite point in tim e i.e. the switching off of the
noise generator, it is possible to average the spectra for a number of reverber­
ation decays at identical points in tim e. For example, suppose 4 0 decays are
measured w ith a read-in interval of 44 ms. A ll the 40 spectra recorded
4 4 ms after the generator is sw itched off are averaged to give one spectrum,
all the 40 spectra recorded 88 ms after the sw itching off of the generator are
averaged to give a second spectrum and so on. These 40 averaged spectra
can then be plotted on an amplitude-frequency-time landscape (see
F ig.6 .6). In this averaged landscape, the points making up the decay curves
in each frequency channel are accurately defined. Therefore by displaying

Desk to p c a lc u la to r
D igital Frequency
A n a lyze r 2131

Noise Generator
1405

c
a • Remote Enable (R E N )
• • •
i>
Power A m p lifie r
2706

770333

Fig. 6.5. Autom atic arrangement - Digital frequency analyser/calculator

method
one of these decay curves on the screen of the 2 1 3 1 , the presence of two or
more slopes can easily be detected and the corresponding reverberation
times calculated. For more detailed information on this method, reference
should be made to B & K Technical Review No.2-1 977.

Fig. 6.6. Representation o f reverberation as a 3-D landscape where the axes

are amplitude, frequency and time

6.3. SO UND DISTRIBUTION

Information about the way in w hich the sound pressure level varies w ith po­
sition w ithin a room is of particular interest to the designers and constructors
of theatres, concert halls and other auditoria. In these buildings, it is vitally
important that the sound be distributed as uniform ly as possible over the area
occupied by the audience.

129
Design Stage — Model Techniques

The effect of a particular design on the sound distribution can be investi­

gated by using a small scale model of the proposed building. It is theoretically
possible to perform in an acoustic model all acoustic measurements which
are usually made in the full sized buildings e.g. reverberation time, sound dis­
tribution etc. However, there are a number of practical problems to be solved
before the model can be realised. Three major problems are the selecting of
suitable transducers w ith acceptable sensitivity and frequency response, the
modelling of the absorption coefficient for the w alls and the modelling of the
absorption coefficient fo r the air. In spite of difficulties acoustic modelling has
been used successfully as a design tool on many occasions e.g. Sydney Op­
era House, Edinburgh Opera House (Ref.8 ). The frequency of the excitation
noise in model experiments should be increased in the same proportion as
the model has been scaled down. If, for example, a model of a concert hall is
to be made at one tenth of the fu ll scale then all frequencies should be in­
creased by a factor of ten. This frequency transform ation can be obtained by
recording the excitation noise on a tape recorder then playing the tape back
at ten times the recording speed. Alternatively, sound sources rich in high fre ­
quency content can be used e.g. the discharging of an electrical spark or an
ultrasonic whistle. In certain circumstances the transm itting transducer can
be a condenser microphone (Ref.31).

Having produced the required sound field w ithin the model, a small m icro­
phone or a probe microphone is used as a receiver. (Small microphones are
used for modelling purposes because they have good high frequency re­
sponse and have little effect on the sound field.) The signal from the receiv­
ing microphone is then amplified and can then be analysed in the usual m an­
ner either before or after the reverse frequency transform ation is performed.
By changing the position of the receiving microphone in the model room, the
sound distribution w ith in the model can be investigated (Fig.6 .7).

Fig.6. 7. Investigating the distribution o f sound w ith in a model o f a concert

h a ll

130
 Model techniques can also be used to investigate, to a certain extent, such
subjective qualities as speech intelligibility, room colouration etc. Speech or
music is first recorded in an anechoic room and then played at ten times the
normal speed into the one tenth scale model room. The sound w ithin the
model room is then recorded and played back after frequency transform ation
to a listener via headphones. Headphones are generally more linear in their
frequency response than loudspeakers therefore the use of headphones is to
be preferred.

Existing Room — Measurement Technique

In an existing room, the sound distribution can be measured directly by in­

stalling a sound source in the room and employing a sound level meter to
measure the sound pressure levels at various positions in the room (Fig.6 .8).

Sound 79

A Power
Source
4205

78
80
Sound
• O • • Source
placed
on
77 p u lp it

Precision
Sound Level
Meter and
Octave 77 76
Analyzer
2215
771215

Fig. 6.8. Investigating the sound distribution in an existing room: sound pres­
sure levels in dB w ith in a church w ith a sound source placed on the
p ulpit

The source should be placed in the most probable positions of the sound
sources w hen the room is in normal use i.e. on the stage of a theatre, on the
pulpit of a church, on the speaker's platform in a lecture theatre. From
these measurements it can be decided w hether further acoustical treatment
by means of reflecting surfaces and sound diffusers is necessary to obtain
the required uniform distribution. The sound distribution sometimes has to be
assisted by means of a loudspeaker system.

An example of the use of sound diffusers in a building already in existance

can be seen in the Albert Hall in London. Here, huge "flying saucer" shaped

131
diffusers have been suspended beneath the dome of the concert hall thus pre­
venting sound energy from entering into the dome cavity and returning into
the main volume of the hall at a slightly later time and producing an irritating
echo.

Loudspeaker systems can be found in most large churches and cathedrals

where "colum ns” of loudspeakers are used to direct the sound of the orator's
voice towards the back of the congregation. A column loudspeaker consists of
several ordinary loudspeakers placed one above the other. Such a system pro­
duces a narrow beam of sound in the vertical plane and a wide beam of
sound in the horizontal plane. The main beam of the column is directed at
the congregation w hich absorbs most of the sound energy. Little energy is ra­
diated to the rest of the room so that the direct sound level is maintained at a
high level while the reverberant sound level is kept low.

6 .4. SO UN D ABSORPTION

The sound absorption coefficient, a , of an acoustic material is d iffic u lt to

measure precisely because a depends on the manner in which the material
is installed, whether the tests are carried out on large or small samples, on
the angle of incidence of the sound on the sample, upon the characteristics
of the room etc. The absorption coefficient claimed by the m anufacturer of a
particular acoustic m aterial should therefore be looked upon as an average to
be used by the acoustic engineer in estimating the am ount of sound absorp­
tion required.

Several methods have been developed for the determ ination of sound ab­
sorption coefficients each method giving a different value of Oi for the same
test m aterial. As of yet, no reliable method has been devised for relating the
results of one method w ith the results of another. Each method, however,
has its use.

Three methods w ill be described here, namely the reverberation room

method, the standing wave method and the tone burst method.

Reverberation Room Method

This method entails measuring the change in reverberation time of a spe­

cially constructed reverberation room w hen samples of sound absorbers are
introduced into the room. The equivalent absorption area of separate objects
such as furniture, people and functional absorbers may also be measured by
this method.

If the reverberation tim e of the room at a certain frequency (frequency

132
band) before the introduction of the sample was T t and after the introduc­
tion, T2 , then the equivalent absorption area, A, of the empty reverberation
room is increased by AA w hich can be calculated from

( 6 . 1)

where V = volume of the room

c= velocity of sound under specified measuring conditions (tempe­
rature, humidity etc.)

For a plane absorber mounted on the w all, floor or ceiling of the room the
absorption coefficient of the specimen may be calculated from

Due to diffraction which can occur at the edges of the test specimen, the
equivalent absorption area determined by th is method is not always directly
related to the area of the specimen. It is possible that the values for absorp­
tion coefficients resulting from this test method are greater than unity.

For the case of a poor absorber (e.g. plaster) it may be necessary to con­
sider that A A is actually the difference between the equivalent absorption
area of the specimen and th a t of the portion of the wall or flo o r covered so
that eqn.6.2 has to be corrected to

AA
(6.3)

where crw = absorption coefficient of the w all or floor covered

The ISO has published a Recommendation entitled "IS O /R 3 5 4 Measure­

ment of absorption coefficients in a reverberation room'' w hich is intended to
promote uniform ity in the methods of measuring . In this Recommendation
it is stated that a suitable reverberation room should have a m inim um volume
of 180 m3 and that the w alls should be hard and smooth and no two walls
should be parallel, i.e. the room should be of irregular shape (Fig.6.9).

Any of the instrumentation arrangements mentioned under the heading re­

verberation tim e may be used to determine cx according to this method.

133
 162336/1

Fig. 6.9. A reverberation room suitable fo r the measurement o f absorption co­

efficients. A sample of the test material o f area 10m 2 is shown
m ounted on the w all

Standing Wave Method

In this method a loudspeaker is used as a sound source to produce stand­

ing acoustic waves in a tube of uniform cross section w hich is terminated by
a sample of the material to be investigated. If the tube w ere terminated w ith
a nearly perfect reflector, such as a steel plate, then the standing wave pres­
sure pattern established in the tube by the interference of the incident and
reflected waves would have the form of th a t shown to the left in Fig. 6 . 10(a).
W hen the tube is term inated w ith an absorptive m aterial then the varying
phase relationship between the incident and reflected waves causes the
acoustic impedance to vary from point to point along the pipe producing a
standing wave pressure pattern like that shown in Fig.6 . 10(b).

134
 Tube terminated by a perfect reflector Tube terminated by sound absorbing material

780254

Fig. 6 . 70. Standing wave pressure patterns show ing the effect o f term inating
the wave tube w ith sound absorbing m aterial

To determ ine the absorption coefficient of the sample, the relationship be­
tween the sound pressure maxima and m inima must first be measured. This
quantity is known as the standing wave ratio and is defined by

SW R = (6.4)
Pmin B

„ I SWR — 1 \ 2
M s T w T 7 ) (65)

where SWR = standing wave ratio

Pmax = maximum sound pressure
P m in = minimum sound pressure
A = amplitude of the incident wave
B = amplitude of the reflected wave

As the absorption coefficient of the sample is defined as the ratio between

the energy absorbed by the sample to the total energy incident on the sample
and as the energy is proportional to the square of the sound pressure then

a= 7- (I)' (6 .6,

It is therefore particularly easy to determ ine the absorption coefficient for

normal incidence using this method. The Standing Wave Apparatus Type
4002 is specially designed for this type of measurement. The 4 0 0 2 consists
of a long tube w ith a loudspeaker mounted at one end. A t the other end is
mounted either a sound reflecting metal plate or a circular sample of the
material to be tested. The loudspeaker emits a sound wave down the tube
and this wave is reflected from either the metal plate term ination or the test
sample. The pressure maxima and minima w ith in the wave tube can be de­
tected by means of a microphone probe tube which is led through an axial

135
 Fig. 6. 7 7. The Standing Wave Apparatus Type 4 0 0 2 and accessories

hole in the loudspeaker. The probe tube is supported w ith in the wave tube by
a small gliding carriage. The other end of the probe tube is connected to a
microphone car which can be moved back and forth over a graduated track.
As the measuring method requires plane waves then the diameter of the sam ­
ple m ust not be greater than about half the wavelength at the frequency of in ­
terest. To enable measurements to be performed over a relatively wide fre ­
quency range, the 4 0 0 2 is supplied w ith tw o wave tubes of different diam e­
ters. The larger one (of 100 mm diameter) is intended for the frequency
range of 9 0 Hz to 1 80 0 Hz and the sm aller one (of 30 mm diameter) for the
range 8 0 0 Hz to 6500 Hz. Each tube is supplied w ith three sample holders of
different design; two of fixed depth and one of adjustable depth.

To determ ine a of a particular material, the 400 2 has to be used in conjunc­

tion w ith a generator, am plifier and a filte r or with the Heterodyne Analyzer
Type 2 0 1 0 which contains the generator, filte r and am plifier in the one in ­
strum ent. A circular disc of the absorbing material is attached to the end of
the wave tube by means of one of the sample holders supplied. Plane waves
are produced in the tube by the loudspeaker which is fed by the generator.
The positions of the maxima and minima produced w ith in the tube are deter­
mined by moving the probe microphone u ntil a maximum (or minimum) deflec-

136
 Measuring A m p lifie r 2 606

Fig. 6. 12. Measurem ent of absorption coefficient employing the Standing

Wave Apparatus Type 4 0 0 2

tion is obtained on the measuring amplifier. The position of the car can then
be read from the graduated track. If the special scale Type SA 0 0 5 4 be fitted
to the measuring amplifier then the absorption coefficient may be read di­
rectly. To enable the pressure maxima and minima to be determined more ac­
curately, a Measuring Am plifier Type 2606 can be made to selectively am­
plify the output of the microphone. This is done by connecting the 2 6 0 6 to a
slave filte r which is automatically tuned from the 1 0 2 3 or by using a Hetero­
dyne Analyzer Type 2010 as m entioned previously. This arrangement m ini­
mises the effects of noise and harmonic distortion. Such precautions are
especially necessary when the m aterial under test has a small coefficient of
absorption because in this case the reflected wave is almost of the same
magnitude as the incident wave; hence the sound pressures at the minima
become very small.

Using the Standing Wave Apparatus Type 4 0 0 2 , it is a straightforward mat­

ter to determine the acoustic impedance of an acoustic material. Such infor­
mation is necessary when the boundary conditions of a room's lim iting sur­
faces have to be inserted into the wave equation to investigate such acoustic
problems as flu tte r echoes and sim ilar phenomena. An acoustic material
whose acoustic impedance is real (i.e. where there is no phase difference be­
tw een the pressure and the particle velocity at th e surface of the material)
w ill have a pressure maximum precisely at the surface of the m aterial and a
pressure m inim um exactly a quarter wavelength away. Hard reflective sur­
faces have acoustic impedances approaching the value of zero. If the sample
possesses a complex acoustic impedance then the reflected wave is out of
phase w ith the incident wave. The phase angle and thus the real and imagi-

137
nary parts of the complex impedance can be determ ined by measuring the
standing wave ratio and the distance, d, by which the first pressure minimum
or node is displaced w ith respect to the position the node would have occup­
ied if the wave tube had been term inated w ith a perfectly reflecting surface.
The complex impedance can then be obtained by calculation from these two
quantities or, a far neater method, these two quantities can be entered on a
Sm ith Chart and the impedance read directly. (N.B. the Smith Chart is a nom­
ographic device w idely used in the field of electronics to determine th e imped­
ance of transmission lines.)

Tone Burst Method

This method enables the absorption coefficient to be determined fo r various

angles of incidence of the sound energy. An advantage of the method is that
no special reverberation room is required for the tests.

The principle of the method is straightforward (Ref.37). A tone burst is pro­

duced by gating a sinosoidal signal. The tone burst is emitted from a loud­
speaker into the test room and the pressure signal is received by a micro­
phone at a known distance, x, from the loudspeaker (Fig.6 .1 3). The
loudspeaker is then aimed at the test specimen at the desired angle of inci­
dence such that the distance between the loudspeaker and the point o f reflec­
tion is x /2 . Alm ost any type of loudspeaker may be employed as a highly di­
rective loudspeaker is not required fo r this method. The microphone is posi­
tioned to receive the reflected sound, the total path length from loudspeaker,
to specimen, to microphone being x.

By comparing the sound pressures (determined from the sound pressure le­
vels measured by the microphone) fo r the direct and the reflected sound, the
reflection coefficient rgf, of the specimen at that particular frequency and at
that particular angle of incidence can be calculated. The absorption coeffi­
cient cxgf, can then be determined from

(6.7)

The reflection coefficient is the ratio between the intensity of the reflected
sound to the intensity of the direct sound i.e. the ratio of the square of the
reflected sound pressure to the square of the direct sound pressure.

Using the automatic sine sweep facility of the sine generator, the sound
pressure level due to the reflected sound from a w a ll relative to th a t due to
the direct sound from the loudspeaker can be measured as a function of fre­
quency. The absorption coefficient as determined from these results is shown
in Fig.6.14.

138
 ■ß-.m i
•<D _

A m p lifie r
K ]
2606

Oscilloscope
4714

Level Recorder A m p lifie r ............... -

2307 2706

Fig. 6.13. Tone burst method o f m easuring absorption coefficients

Sound Pressure

Frequency Hz

A b sorp tion
C o efficient
a
1

0,5

1000 2000 5000 10000 20000

Frequency Hz

Fig. 6.14. The absorption coefficient o f a w all as function of frequency for an

angle of incidence of 3 0 ° as determined from the differences be­
tween the direct sound and the reflected sound

139
6 .5 . SOUND INSULATION

Sound measurements can be conveniently divided into airborne and impact

sound insulation. These two headings can be subdivided into field and labora­
tory measurements. The practical acoustician w ill be mostly involved in field
measurements of sound insulation as the laboratory measurements require
special facilities w hich, for economic reasons, are usually the perogative of
the large institutions such as government owned research centres. The basic
instrum entation in sound insulation measurements requires a sound source
(loudspeaker or standard impact machine) placed in the em itting or source
room and a microphone in conjunction w ith filte rs and an am plifier to mea­
sure the sound pressure levels in the receiving and source rooms.

In each frequency band of interest the sound pressure levels produced by

the sound source should be at least 5 dB above the background noise level in
that band. The average sound pressure level in the source and receiving
rooms should be determined either by using a num ber of fixed microphone po­
sitions or a single microphone whose output is integrated w hile the micro­
phone is swept over a particular path to obtain the spatial average.

Airborne Sound Insulation

Field Measurements

The three battery operated instrum ents shown in Fig.6.15 form a compact
set suitable for octave band insulation measurements on walls, w indow s, do­
ors etc. and also for reverberation measurements (see section Reverberation
Time). For measurements in third octave bands e ithe r the Precision Sound Le­
vel Meter Type 2 2 0 3 or 2209 should be employed in conjunction w ith the
Third Octave Filter Set Type 1 6 1 6 . Knowing the sound pressure levels pro­
duced in each room, the equivalent absorption area of the receiving room (de­
term ined from reverberation measurements) and the area of the dividing wall
then the sound reduction index of the dividing w a ll can be calculated accord­
ing to Eqn.3.2. In field measurements, however, it is usually the apparent
sound reduction index, as calculated from Eqn.3.4, that is of interest. A preci­
sion Integrating Sound Level M eter Type 221 8 can be used together w ith the
Rotating Microphone Boom Type 3 9 2 3 to determ ine the average sound pres­
sure level over a circular path. F ig.6 .1 6 shows such an arrangement used for
determ ining the sound insulation of an external w indow . In blocks of flats or
sim ilar buildings, it may be w orth w h ile to record on a tape recorder the
sound pressure levels in several rooms simultaneously (Fig.6.17). The record­
ings can then be analysed later in the laboratory to yield the sound reduction
index of the w alls between the source room and the receiving rooms using,
for example the instrumentation show n in Fig.6 .1 8.

140
Fig.6.15. Field measurements employing the Sound Power Source Type
4205

Fig. 6.16. Field measurements employing a rotating microphone boom and an

integrating sound level meter

141
 B represents a Precision Sound Level Meter Type 2206
W 770385

Fig. 6.1 7. Recording the sound pressure levels in several rooms simultane­
ously

142
Laboratory Measurements

Fig. 6.19. Sound insulation measurements in a laboratory sound transmission

suite

Laboratory measurements of sound insulation are usually performed in a

specially constructed transm ission suite w here as far as is practically possi­
ble the only way for airborne sound (or impact sound w hen dealing w ith im ­

143
pact noise) to be transm itted from the source to the receiving room is through
the wall or partition (or floor) under test. The most common method of isolat­
ing the tw o test rooms from each other is to build two completely independ­
ent structures as shown in Fig.6.19. As the flanking transm ission which oc­
curs in this structure is very small relative to the direct transmission, the
sound insulation afforded by a wall or partition as determined in this way in
the laboratory is always som ewhat greater than that experienced in practice.

The arrangement shown in Fig.6.19 can be used to measure the sound in ­

sulation in third octave bands employing several microphone positions. Filter
sets are used to lim it the signals to both the sound source and the measuring
amplifier to third octave bands. This means that:

1) a higher sound pressure level in the th ird octave band of interest can be
obtained from a given loudspeaker than if a broad band noise signal were
used.

2) the effect of background noise on the measured sound pressure level is

reduced.

The filte r sets are switched in unison from one filter band to the next by an
electrical signal from the level recorder. The curves in Fig.6 .20 show typical
sound level differences between the two rooms of a transm ission suite.

* Bruel & Kjær Brüel & Kjær Brüel & Kjær

B r üe l & K j æ r Potentiometer R ange:_5Q_dB Rectifier:— Q£--------Lower Lim.

Copenhagen

Bimii
Measuring Object:

Sound level
difference _
between
two offices-

Rec. No.:--------------
Date: 2 1 / 9 / 7 8 ___
S ig n .:M .M .______
3347:

Weight. N e tw .:__
Ref. Level:4 5 dB
Third Octave Band 1g
QP 1152C en,er Frequency^ 2,5

Fig. 6.20. Typical sound level differences between two rooms. Note that the
D, A, B, C & Lin. levels have been attenuated by 2 0 dB

A rotating boom can be employed to sweep the microphone around a circu­

lar path to obtain the spatial averaged sound pressure level. The sound pres­

144
 sure detected by the microphone can be integrated over a tim e interval equal
to the duration of one complete microphone revolution by means of either the
Noise Level Analyser Type 4 4 2 6 or the Precision Integrating Sound Level M e ­
ter Type 2 21 8. When using the 4426 fo r this application, a separate m icro­
phone power supply is required.

The measurement of the sound reduction index of a building element can

be automated to a large extent by employing a digital frequency analyser in
conjunction w ith a desk top calculator (Fig.6.21).

Digital Frequency Desktop C alculator

A n a ly z e r 2131
____________________________ 780257

Fig.6.21. Laboratory measurements em ploying a digital frequency analyser

and a desk-top calculator

The signal from the rotating microphone is integrated and analysed into
third octave bands by the Digital Frequency Analyser Type 2 1 3 1 . The spectra
of the sound fields in both rooms of the transmission suite can be sim ultane­
ously displayed in the form of third octave band sound pressure levels on the
screen of the 2131. By using a programmable desk top calculator, the re­
quired sound reduction index (or apparent reduction index) can be evaluated
in a trice from the data stored in the 2 13 1.

The necessary programme is available from Briiel & Kjaer in the Acoustic
Programme Packages BZ 001 1 or BZ 001 2.

145
Im pact Sound Insulation

Field measurements

For both field and laboratory measurements of impact sound insulation, a

standardized tapping machine is usually employed (i.e. one w hich complies
w ith ISO 1 40). The measuring technique for impact sound insulation is sim i­
lar to that for airborne sound insulation. The Tapping Machine Type 3204 is
used to excite the partition (usually the floor) and the average sound pressure
levels are measured in octave or in third octave bands in the receiving room.
The spatial average of the sound pressure levels can be determined by either
using a number of microphone positions (see Fig.6.22) or by using a Preci­
sion Integrating Sound Level Meter Type 2 2 1 8 and integrating the output of
a microphone which is continuously swept over a prescribed path (Fig.6.23).

770377

Fig. 6.22. Field measurements o f impact sound insulation using a number o f

microphone positions

146
Fig. 6.23. Field measurements of impact sound insulation using an integ ra t­
ing sound level m eter and a rotating microphone boom

Laboratory Measurements

Two arrangements suitable for the laboratory are shown in Fig.6 .24 and
6.25. In Fig.6 .24 the level recorder automatically switches the filters from
one filter band to the next. A typical spectrogram recorded by this method is
illustrated in the diagram.

147
 Y7ZZZZZZZZZZZZZZZZZZZZZZZZA
Source room

Measuring A m p lifie r
2608
Tapping M achine
3204

I; B25:l Q 9
•O -E * * 0«
Receiving room
F ilte r Set 1618

M icrophone
4 1 3 4 '- ^ ^

P r e a m p l if ie r '^ ^ .
2619 J

Filter
^ ^ r i p o d U A 0049 switch
Level Recorder control
2307

Typical Spectrogram

Fig. 6.24. Laboratory measurement of im pact sound insulation using several

microphone positions with a typical octave band spectrogram

148
 770395

Fig. 6.25. Laboratory measuremnet o f impact sound insulation employing a ro­

tating m icrophone and a d ig ita l frequency analyser

6 .5 . SOUND POWER

It is becoming more and more common to rate the sound output of ma­
chines and domestic appliances in term s of sound power. The measurement
of the sound power output is being introduced more and more as a part of
the quality control testing of the machines. It is not, however, a sim ple mat­
ter to measure sound power. This statement is confirm ed by a study of the 6
new ISO publications on the subject. Special test facilities are required such
as a reverberant room, a semi-anechoic room or an anechoic room and many
sound pressure measurements have to be performed before the sound power
can be calculated. The rating values are valid fo r factory tested new ma­
chines but after installation or prolonged use the sound output of the ma­
chines may w ell change. It can be important for the acoustician to know
w hether the machines installed in a particular building fu lfill their rated val­
ues.

149
 (The Sound Power Source T y p e 4205 y/ / / / / / / / 7 / Z / / / 7 / / / / / / / 7 / / / / / / / / / / / / / / 7 /
' / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /
consists o f both G enerator and Sound
Source HP 1001) Sound Source Precision Sound Level / /
M ete r and Octave A n a ly z e r /
/
/
/

/
/
:__ J | /

: s * :
* M achine ^/
/ S S S S S // // / / ' / / / / / / / / / / / / / / / / / /
/ / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / ¿ A y/

771190

Fig. 6.26. Sound pow er measurements employing the Sound Pow er Source
Type 4 2 0 5

The rating of a machine in situ can, in many cases, be determined by em­

ploying the Sound Power Source Type 4205. This instrument gives a rapid
and accurate value of the sound power level w hich can be read directly from
a meter.

Further inform ation about the various measurement methods both in the la­
boratory and in situ can be found in Ref.28 and 35.

6 .6 . ROOM M O D ES

Room modes and their effects e.g. the "co lou ra tion " they impose on the
sound output of loudspeakers, can be investigated using the instrumentation
shown in Fig.6 .2 7 . More precisely, the instrum entation measures the irregu­
larity of the transm ission of pure tones from a source located at one point in
a room to a microphone located at another point. If a loudspeaker is placed in
one corner of the room and the microphone is placed in the diagonally oppo­
site corner and a relatively slow sweep rate for the pure tone is used then
practically all the room modes w ill be excited. Fig.6 .2 8 shows a typical exam­
ple of the transm ission of sound for such loudspeaker and microphone posi­
tions for a room of 2 ,5 x 3 x 7 m when the pure tone was slowly increased
from 20 Hz to 1 0 0 Hz. The normal modes of vibration corresponding to the
various resonant frequencies are specified by th e numbers in parenthesis
above the resonant peaks. Repeating the measurement for the same loud­
speaker and microphone in an anechoic room w ou ld give a response similar
to the dashed curve in Fig.6.28. The difference between these tw o curves is
the colouration th a t the room places on the free field loudspeaker response.

150
 P re a m p lifie r 2619
M ic ro p h o n e 4 1 3 4 - ^
Measuring A m p lifie r
2606
Power A m p lifie r
2706
I.:

G ; • 3
• # •
filt e r tu n in g signals

• " - r— o» •*- O» —
, 0 0 •®•<§>
Sine Generator Heterodyne
1023 Slave F ilte r
2020

frequency sweep control

Level Recorder
2307

Fig.6.27. Investigation o f modes w ith in a room

171026

Fig. 6.28. Response o f a loudspeaker in a room o f dimensions 2 ,5 x 3 x 7 m

over the frequency range 2 0 Hz to 100 Hz

151
 Particular peculiarities of the room e.g. prominent resonances and marked
unresponsiveness over a particular frequency range, can easily be seen from
a record of the microphone's output.

6 .7 . DIRECTIVITY OF NOISE SOURCES

A knowledge of the directivity of noises sources is of importance, fo r exam­

ple, in the planning of workshops and factory halls, w here the correct orienta­
tion of the machines and the correct positioning of sound absorbing materials
can lead to substantial reduction in the general noise levels.

The directivity of a noise source w hich is situated in a building can be deter­

m ined by measuring the sound pressure levels produced at a given distance
from the source using a sound level meter. Filters may be used to obtain the
directivity for various octave or th ird octave bands. This method yields reliable
results providing the principal contributor to the sound pressure level is the di­
rect and not the reverberant field. If the source can be moved then the direc­
tiv ity can be determined in an anechoic room. Small sources can be mounted
on the Turntable Type 3922 and the receiving microphone can be placed in a
fixed position w hile the source is rotated. The results of the measurement
can be plotted directly onto polar graph paper by employing the Level Recor­
der Type 2307.

6 .8 . VENTILATION A N D SERVICE SYSTEM NOISE

The noise produced by ventilation and service systems (gas and oil fired
burners, waste disposal chutes, w a te r cisterns etc.) in a building can be mea­
sured using a sound level meter. For measurements in octave bands the Pre­
cision Sound Level Meter and Octave Analyser Type 221 5 can be used. For
th ird octave analysis, the Filter Set Type 1616 can be used w ith e ither the
Sound Level M eter Type 2209 or Type 2203. For measuring sound pressure
levels in regions of high turbulence, for example w ith in airducts, a 1 / 2 inch
microphone can be used fitted w ith the Turbulence Screen Type UA 0 4 3 6 .

For laboratory methods of measuring the noise em itted by valves, pipes, w a­

ter heating appliances etc. reference should be made to the ISO publication
3 8 2 2 /1 entitled, "Acoustics - Laboratory tests on noise emission by appli­
ances and equipment used in w a te r supply installations". These measure­
m ents may be useful to predict the noise due to plumbing in situ.

152
6.9. VIBRATION M EASUREM ENT

Locating and monitoring vibration

For locating and m onitoring vibration, the Vibration M eter Type 2511 is rec­
ommended. The transducer can either be attached w ith wax or a magnet to
the structure under test or the transducer can be held in the hand as a
probe.

This compact, battery operated instrum ent can be combined w ith the T un­
able Band Pass Filter Type 1621 in a carrying case to produce a self con­
tained vibration analysis system known as the Portable Vibration Analyser
Type 3 5 1 3 . Adding the Portable Level Recorded Type 2 3 0 6 to the system
greatly increases the possibilities e.g. enables automatic recordings of fre ­
quency analyses to be made.

Fig. 6 .2 9 . M easurem ent o f sound and vibration due to run nin g water

An example of vibration monitoring is shown in Fig.6 .3 0 . Here a machine

has been mounted on anti-vibration mountings. The improvement in vibration
isolation can be measured by monitoring the vibration levels on the base of
the machine and on the structure of the building. The sound pressure levels

153
produced by the machine are monitored sim ultenously by a sound level me­
ter. The data can be recorded on a tape recorder and then analysed later in
the laboratory.

Precision Sound Level Meter

2206
Measuring

C o n d itio n in g
A m p lifie r
2635
Accelerom eter
4368
Tape Recorder
700 3

Accelerom eter C o n ditio ning

830 6 A m p lifie r
2635
•••
*« ♦
J■ • •

/ / / / / / / / / / / / / 7 /
Analysing

Tape Recorder Frequency Analyzer

7003 2120
Level Recorder
2307
-‘ S‘ 9 1

filte r sw itch c o n tro l

Fig. 6.30. Recording and analysing the vibration isolation provided by m oun­
ting a machine on resilient supports

Complete sets are available in robust carrying cases w hich contain all the
necessary equipment for making both sound and vibration measurements.
Fig.6.31 shows a typical set containing a sound level meter, a filter set, a
microphone, accelerometers and all the accessories.

154
 Fig. 6 .3 1 . The Sound and Vibration Set Type 350 7

Two mains operated systems for the detailed narrow band analyses of vibra­
tion signals are shown in Figs.6 .3 2 and 6.33. The vibration levels of a sur­
face can be expressed as a velocity, an acceleration or a displacement. When
measuring the level of vibration of a surface w ith in a room, it is sometimes
useful to determine the average surface velocity levels of the surface and to
express the results in decibels using the equation

<■B <6.8,

w here Lv = average surface velocity level

, v 2, ...v are the rms normal surface velocities at n different
positions on the wall or ceiling
vQ = 10 ~_9 m .s.—1 the reference velocity

155
 Fig. 6.32. N arrow band vibration analysis employing heterodyne analyser

X - V Recorder
2308

771211

Fig. 6.33. N arrow band analysis employing a spectrum analyser

156
Flanking Transmission

The instrumentation described in the foregoing section may be used to mea­

sure the r.m.s. normal surface velocities of a flanking surface in the receiving
room at several different positions (for a medium sized room about six acceler­
ometer positions on each surface are sufficient). The amount of energy radi­
ated from this surface into the room can be estimated from equation 3 .8 , i.e.

Wk = p c S kVZ<jk ( 6 .9 )

The quantity W k can then be compared to the am ount of energy w hich is

radiated into the receiving room from the dividing room via the common wall
thus giving a measure of the flanking transmission via that particular flanking
surface (Fig.6 .34).

Flanking Wall 780259

Fig. 6.34. D eterm ining the am ount of ftanking transmission between two
rooms by measuring the rms normal surface velocities o f the build­
ing elements

Loss Factor

The loss factors of the various building elements may be measured by deter­
m ining the reverberation time of the wall. A vibration exciter forces the wall
into vibration; the exciter is abruptly stopped and the decay of the level of ac­
celeration or velocity of the w all is monitored by an accelerometer and the de­
cay in level is drawn by a level recorder (Fig.6 .35). From the decay curve a re­
verberation time can be determined and hence the loss factor by using

157
w here 77 = loss factor
f = octave band or third octave band centre frequency
T = reverberation time in the octave or th ird octave band

Wall, Floor

Power A m p lifie r Noise G enerator

2706 1405
V ib ra tio n E xciter 0• • •
4809 . * • . • r%-. •
# <ê
i _____ o

H Level Recorder
2307
Charge A m p lifie r F ilte r Set
2635 1618
Accelerom eter
/ 4367
fTSTTH
•O -- i

Fig. 6 .35 . Determining the loss factor o f a building element

N arrow Band X -Y Recorder

Spectrum A n alyzer 230 8
2031

Fig. 6.36. Capture and analysis o f a shock signal using a Narrow Band Spec­
trum Analyzer Type 2 0 3 1 and an X— Y Recorder Type 2 3 0 8

Transmission of Shock

W hen investigating the transm ission of shocks through a building a simple

but very effective way of imparting a shock to the building structure is to let a
sack of sand fall from a height onto the floor. The shock produced can be dis­
played, analysed and recorded using the instrum entation shown in F ig.6.36.

Mechanical Mobility

W hen a source of structure borne noise has been found in a building and
when further damping of the source would be of no avail, there rem ains the
possibility of reducing the transm ission sound by damping or closing the
transmission paths. The efficacity of a particular transm ission path as a func­
tion of frequency can be determined by measuring the mechanical m obility of
the structure (Ref.23 and 33). The mechanical m obility is defined as the ratio
of the vibrating force applied to a certain point on a structure to the resulting
velocity of vibration. A maximum in the mobility response curve corresponds
to a natural mode of vibration of the structure and therefore to a low trans­
mission loss.

A typical instrum entation arrangement for measuring mechanical mobility

is shown in Fig.6 .37.

C o n d itio n in g

V ib ra tio n 2708 2307

Exciter
4802T + 4819
__________ 780322

Fig. 6 .3 7. Measurement of m echanical m obility

159
 APPENDIX

DERIVATION OF THE NORMAL M O D E EQUATION FOR A RECTANG U­

LAR ROOM

Consider a rectangular room w ith hard, smooth, parallel walls. The room di­
m ensions are lx, l y and lz (Fig.A. 1)

Fig.A.1 Rectangular room o f dimensions /x, ly and lz

The general expression for a plane sinusoidal wave may be w ritte n in the
form

p = A e j(e>t - k x m ^ k y y - k z z ) (A. 1 )
For this equation to satisfy the general wave equation
 then the wave numbers kx, ky and kz must satisfy

* = 7 = A 2+ V + K 2 (A. 3)

which can be verified by substituting eqn. A.1 in eqn. A .2. The quantities
k x / k , k y / k and k z / k represent the direction cosines made by the propagation
of the waves w ith respect to the x, y and z axes. One can replace one, two or
all three negative signs in eqn. A.1 to obtain seven additional equations simi­
lar to eqn.A.1 all having identical values of k x , k y and k z . This array of eight
wave equations represents the fam ily of waves generated by th e original
wave as it undergoes reflections at the boundaries of the room.

The particle velocities in the three principal directions u, v, and w , must be

determined in term s of the pressure, p, before the boundary conditions can
be applied to the problem. For a sinusoidal acoustic wave

ôp du
- T x = p W = Ja)f,u {AA)

., , — 1 ôp
therefore u = -— . (A.5)
J<x>p ox ' '

Sim ilarly v = - —- (A.6)

jw p dy

and w = ——- . (A.7)

jw p dz

As the walls of the room are assumed to be rigid there can be no normal
velocity component of the air particles near to any w all. These boundary con­
ditions may be stated as

u = O for x = O and x = lx

v = O for y = 0 and y = l y

w = O for z = O and z = lz

Applying the boundary conditions at x = 0, y = 0 and z = 0 to the respective

equations for the particle velocities in the equations of the form of eqn.A.1.
yields the undamped standing wave equation

p = P (cos kxx ■ cos kYy ■ cos k zz)e'"n (A.8)

161
w here P is the magnitude of the pressure. Equation A .8 . shows that pressure
antinodes existat the w all surfaces i.e. where x = 0, y = 0 and z = 0. Substi­
tu ting eqn.A .8 . into eqn. A .5., the particle velocity in the x-direction be­
comes

k P
u = -A— (sin k .x ■ cos k vy ■ cos k zz ) e '" t (A.9)
Hop

Applying to eqn. A .9. the condition

for x = /x then s i n ( k j x) = 0 (A .10)

or kx = — wher e nx = 0 ,1 ,2 ,3 ... (A .11)

Sim ilarly = ~~7~ w here ny = 0 ,1 ,2 ,3 ... (A .12)

'V

and kz = r-Jj — w here nz = 0 ,1 ,2 ,3 ... (A .13)

'z

From eqns. A .3, A .1 1 , A. 12 and A. 13 one obtains the important expres­

sion

. CO c 2] 1/2
(A. 14)
2n^2

w hich gives the permissable frequencies corresponding to the normal modes

of vibration w ithin the room.

162
 BIBLIOGRAPHY

GENERAL REFERENCES

1. M.J.Crocker & A.J.Price Noise and noise control, CRC press 1 9 7 5

2. M easurem ent o f sound insulation in buildings and of building elements,
ISO/D 140
3. Robin MacKenzie, A uditorium acoustics. Applied Science publishers,
1975
4. Leo L. Beranek, ed., Noise and vibration control, McGraw-Hill 1 971
5. Helmut Schmidt, Schalltechnisches Taschenbuch, VDI Verlag 1 9 7 6
6 . J.D.Webb, ed., Noise control in industry, Sound Research Laboratories
Ltd. 1976
7. Akustik for bygningsteknikere, Lydteknik teknologisk institut, Teknologisk
instituts forlag 1976
8 . Ian Sharland Woods practical guide to noise control. Woods Acoustics
Ltd., 1973
9. D.W.Robinson, ed.. Occupational Hearing Loss, Academic Press 1971
10. R.H.Warring, Handbook of noise and vibration control. Trade and Techni­
cal Press Ltd., 1 970
1 1. A.B.Lawrence, Architectural Acoustics, Applied Science Publishers Ltd.,
1970
12. R.D.Berendt, E.L.R. Corliss and M.S.Ojalvo, Quieting: A pra ctical guide
to noise control, U.S. National Bureau of Standards 1 976
14. Heinrich K uttruff, Room Acoustics, Applied Science Publishers 1 973
15. W illi Furrer and A.Lauber, Raum und Bauakustik Larmabwehr, Birkåuser
1972
16. Rupert Taylor, Noise, Penguin Books, 1975
1 7. Theodore John Schultz, Com m unity Noise Ratings, Applied Science Pu­
blishers Ltd., 1972
18. W illiam Burns, Noise and M an, John Murray, 1 9 68
1 9. Gideon Gerhardsson, Buller, Uggle Bocker, 1 9 7 0
20. Karl D. Kryter, The Effects o f Noise on Man, Academic Press, 1 9 7 0
21. Lyle F. Yerges, Sound, Noise and Vibration Control, Van Nostrand Rein­
hold, 1969

163
22. K.Bolund, On the use o f the integrated impulse response m ethod fo r la­
boratory reverberation measurements, J. of sound and vibration, 56(3),
1978
23. F.J.Fahy and M.E.W estcott, M easurem ent o f flo o r mobility at lo w fre­
quencies in some buildings w ith long floor spans, J. of sound and vibra­
tion, 57(1), 1978
24. L.E.Kinsler, A ustin R. Frey, Fundamentals o f acoustics, John W iley &
Sons, 1 962
25. Morse and Ingard, Theoretical acoustics, M cG raw -H ill, 1968
26. L.Cremer and H .A.M uller, Die wissenschaftlichen Grundlagen der Rau-
makustik, S.Hirzel Verlag, 197 8
27. W .V.M ontone, Guide to the literature on architectural acoustics. Sou­
theast Acoustics Institute, 1978

B & K PUBLICATIONS

28. Acoustic measurements in accordance with ISO standards and recom ­

mendations, B & K publication 1 9 7 8
29. Application of B & K equipment to mechanical vibration and shock meas­
urements, B & K publication 197 2
30. Acoustic measurem ent using the D igital Frequency Analyser Type 2131
w ith a desk-top calculator, B & K application note 1 977
31. Erling Frederiksen, Condenser microphones as sound sources, B&K
Technical Review No. 3 — 1 9 77
32. 3 — D acoustical measurements using gating techniques, B & K applica­
tion notes 1 9 77
33. Torben Licht, Measurement o f lo w level vibrations in buildings, B & K
Technical Review No. 3 — 1 972
34. Erik Rasmussen, Acoustic response of theatres, B & K Technical Review
N o.2 — 1972

35. Acoustic noise measurements, B & K publication 1 978

36. Frequency analysis, B & K publication 1 977

37. Electro-acoustic free field measurements in ordinary rooms using gating

techniques, B & K application notes 1 977
38. Condenser microphones and microphone pream plifiers, B & K publication
1977
39. Piezoelectric accelerometers and vibration pream plifiers, B & 'K publica­
tion 1 978
40. Accelerom eter calibration, B & K publication 1 9 7 8

164
 INDEX

Absorber, acoustic .......................................................................................3 7 , 53

— panel ..............................................................................................................5 5 , 58
— functional ............................................................................................................. 60
— resonator ............................................................................................................. 56
Absorption, coefficient ............................................................................. 3 7 , 38, 39
— effective ................................................................................................................ 39
Accelerometer, calibration ......................................................................................114
— selection of ....................................................................................................... 110
Acoustic Jeaks ............................................................................................................ 78
Acoustic material ................................................................................................58, 59
Acoustic modelling .................................................................................................. 130
Acoustic fault, dead spot .................................................................................. 48, 52
— echo ...................................................................................................................... 48
— flutter echo ......................................................................................................... 48
Acoustics of buildings ................................................................................................ 61
Acoustics of rooms .................................................................................................... 34
Airborne sound insulation .................................................................... 62, 67, 140
A ir conditioning system ............................................................................................82
Am plifier .................................................................................................................... 118
Amplitude .................................................................................................................... 13
Analyser .................................................................................................................... 118
Articulation tests ........................................................................................................ 48
Assisted resonance technique ...........................................................................50, 51
Average ........................................................................................................................ 15

Bending wave ..............................................................................................................68

Building regulations ...................................................................................................91

Calculator ...................................................................................................................121
Calibration, of accelerometers .............................................................................. 114
— of measuring systems ......................................................................... 109, 114
— of microphones ................................................................................................. 109
Ceiling false ................................................................................................................ 77
— suspended .............................................................................................................77
— transmission loss of ...........................................................................................77

165
Characteristic frequency ................................................................................. 4 0 ,4 1
Chart recorder ...........................................................................................................120
Coincidence effect ...................................................................................................... 68
Coincidence dip .......................................................................................................... 68
Concert hall .......................................................................................................... 50, 51
Critical frequency ........................................................................................................68

Damage risk criteria .................................................................................................. 87

Damping .......................................................................................................................70
Dead spots ................................................................................................................... 48
Debating chamber ...................................................................................................... 49
Decibel ......................................................................................................................... 15
D efinition ..............................................................................................................4 4 , 52
Design of rooms and auditoria .................................................................................46
— for music ...............................................................................................................51
— fo r speech .............................................................................................................48
D iffuser ........................................................................................................................35
Dipole ............................................................................................................................22
D irectivity factor .......................................................................................................... 26
D irectivity index .......................................................................................................... 26
Directivity of noise sources .....................................................................................152
D irectivity pattern ....................................................................................................... 24
Discontinuous construction ..................................................................................... 78
D isplacem ent............................................................................................................... 13
Doors ........................................................................................................................... 73
Double leafed elements ............................................................................................70
Duct ...................................................................................................................... 82, 83

Ear ......................................................................................................................... 29, 86

Energy density .............................................................................................................16
Equal loudness c o n to u r..............................................................................................31
Eyring's formula ..........................................................................................................38

Filter ................................................................................................................ 115, 118

— mechanical ............................................................................................... 1 1 0 , 111
Flanking transmission ............................................................................ 64, 7 2 , 157
Floating floor ...............................................................................................................75
Floor-celing element ..................................................................................................74
Flutter echo ................................................................................................................. 48
Frequency .................................................................................................................... 11
Frequency analyser .................................................................................................118
Frequency space .................................................................................................4 4 , 45
Fullness of tone ..........................................................................................................52
Functional absorbers .......................................................................................60, 132

Generator ..........................................................................................................99, 101

166
Geometrical room acoustics 34

Haas effect ..................................................................................................................48

Hearing damage ......................................................................................................... 86
Helmholtz resonator .......................................................................................... 5 1 ,5 6
Hydrophone .............................................................................................................. 105

Impact sound insulation ................................................................................ 65, 146

Impedance, specific acoustic .................................................................................. 14
Intensity ....................................................................................................................... 17
Intensity level ............................................................................................................. 17
Intimacy ....................................................................................................................... 52
Isolated room .............................................................................................................. 78
Isotropic sound source ............................................................................................ 101

Landscape, three dimensional ............................................................................. 129

Longitudinal wave ................................................................................................. 9, 66
Loss factor .................................................................................................... 1 03, 1 57
Loudness ..............................................................................................................3 1 ,5 1
Loudness level ............................................................................................................ 31

Mass law .....................................................................................................................67

Measuring technique .................................................................................................99
Mechanical mobility ................................................................................................159
Microphone, calibration ......................................................................................... 109
— free field ............................................................................................................1 0 4
— pressure ............................................................................................................ 104
—random incidence .............................................................................................. 104
M illington and Sette's form ula ................................................................................39
Mineral wool ....................................................................................................... 58, 59
Mobility ..................................................................................................................... 159
Mode axial ...................................................................................................................41
— oblique .................................................................................................................42
— tangential ............................................................................................................. 42
Model te ch n iq u e s .................................................................................................... 1 30
Monopole .....................................................................................................................20
Musicians' criteria ......................................................................................................53

Noise airborne .............................................................................................................61

— noise ...................................................................................................................123
— random ...............................................................................................................123
— structure borne .................................................................................................... 61
— white .................................................................................................................. 1 23
Noise control .............................................................................................................. 85
Noise criteria ....................................................................................................... 85, 88
Noise dosage .............................................................................................................. 86

167
Noise in the home ...................................................................................................... 90
Noise rating ................................................................................................ 87, 88, 89
Normal mode ............................................................................................................... 4 0
Normal mode equation .................................................................................... 40, 160
Normalised impact sound pressure level ..................................................... 65, 66

Paper loop method ........................................................................................ 126, 127

Particle v e lo c ity ............................................................................................................14
Peak ............................................................................................................................... 15
Percentage articulation index ................................................................................... 48
Period ..........................................................................................................................11
Phon ............................................................................................................31, 32, 33
Pink noise ............................................................................................................... 1 23
Pistol shot method ..................................................................................... 1 2 3 ,1 2 4
Plenum chamber .........................................................................................................82
Portable instrumentation .................................................................. 1 1 4, 11 6, 1 24
— for sound measurement .................................................................................. 114
— for vibration measurement ............................................................................. 116
Preamplifier for accelerometer, selection of ....................................................... 112
— for microphone, selection of .......................................................................... 107

Recorder chart .......................................................................................................... 120

— level .................................................................................................................... 120
— X-Y .......................................................................................................................120
— tape ......................................................................................................................119
Reflection sound rays ................................................................................................3 4
— laws of ................................................................................................................. 34
Reflector .............................................................................................................. 34, 53
Reverberation ..............................................................................................................52
Reverberation room .......................................................................................132, 134
Reverberation time .......................................................................................... 37, 122
— form ulae ...................................................................................................... 38, 39
— "o p tim u m " ........................................................................................................... 4 7
RMS .............................................................................................................................. 15
Room a c o u stics........................................................................................................... 34
Room c o n s ta n t............................................................................................................ 29
Room dead ...................................................................................................................27
— live ......................................................................................................................... 27
— sem i-reverberant................................................................................................. 27
Room modes ..................................................................................................150, 151
Room radius ................................................................................................................ 29
Rotating microphone boom ....................................................141, 145, 147, 149

Sabin .............................................................................................................................37
Sabine's formula ........................................................................................................ 37
Signal generator ...................................................................................................... 102

168
Sone .............................................................................................................................31
Sound absorber ..........................................................................37, 53, 55, 5 8, 6 0
Sound absorption .................................................................................................... 132
Sound frequency of ....................................................................................................11
Sound nature of ............................................................................................................9
— speed of ................................................................................................................12
Sound distribution ..........................................................................................130, 131
Sound field direct .......................................................................................................27
— reverberant ......................................................................................................... 27
Sound generation ............................................................................................. 6 1, 1 00
Sound insulation ................................................................................................ 6 2, 91
Sound insulation characteristic ...............................................................................63
Sound insulation curve .............................................................................................63
Sound level meter ............................................................................................ 33, 115
Sound power .......................................................................................................... 1 4 9
Sound power level .....................................................................................................19
Sound pressure ......................................................................................................... 14
Sound pressure level .................................................................................................18
Sound propagation ............................................................................................... 9, 10
Sound reduction index .............................................................................................. 63
— apparent .............................................................................................................. 65
Sound reflector .......................................................................................................... 53
Sound source practical ............................................................................................ 2 0
— simple ..................................................................................................................2 0
— theoretical ........................................................................................................... 2 0
Source of s o u n d ................................................................................................20, 1 00
Source of vibration ..................................................................................................1 00
Specific acoustic impedance ....................................................................................14
Speech interference level ........................................................................................ 4 8
Speed of sound ...........................................................................................................12
Standing wave ............................................................................................................ 11
Standing wave ratio ............................................................................................... 135

Tape recorder ................................................................................................. 119, 142

Theatre ........................................................................................................................ 5 0
Threshold of feeling ...................................................................................................3 0
— of hearing ........................................................................................................... 3 0
Tone fullness of ......................................................................................................... 52
Tone burst method ..................................................................................................138
Transmissibility .......................................................................................................... 8 0
Transmissibility function ...........................................................................................81
Transmission, flanking ........................................................................... 64, 72, 1 57
Transmission loss ..............................................................................................6 3, 72
Transmission path .............................................................................................. 64, 71
Transmission suite ..................................................................................................143

1 69
Ventilation system .....................................................................................................82
Vibration analyser ...................................................................................................117
Vibration control ....................................................................................................... 80
Vibration isolation .................................................................................................. 154
Vibration is o la to r....................................................................................................... 80
Vibration locating ....................................................................................................153
Vibration m onitoring ...............................................................................................153

Wall outer ...................................................................................................................74

— solid homogeneous impervious ...................................................................... 67
Warbled tone ............................................................................................................102
Wave .............................................................................................................................. 9
Wave, diverging .........................................................................................................10
— elastic ................................................................................................................... 10
— longitudinal ........................................................................................................... 9
— plane .................................................................................................................... 10
— progressive ...........................................................................................................11
— spherical ...............................................................................................................11
— standing ................................................................................................................ 11
Wavelength .................................................................................................................13
Wave terminology .....................................................................................................10
Wave theory ...............................................................................................................39
W eighting network ................................................................................................... 33
W hite noise .................................................................................................... 102, 103
W ind noise ......................................................................................................104, 105
W indow ...................................................................................................................... 74

170
 Bruel & Kjær
DK-2850 NÆRUM, DENM ARK ■ Telephone: + 45 2 80 05 00 ■ Telex: 37316 bruka dk

>y Næ rum Offset