Approach%20to%20Sound%20System%20Design

1
Approach to Sound System Design

• Sound a little bit of Physics
• SPL and sound propagation in free field
• Room Acoustic some useful definitions
• Intelligibility
• Sound System Design some suggestions
• Q A

2

• SOUND is produced by vibrating objects. These
  move air, pushing and pulling from its
  resting state

These small fluctuations in air pressure travel
away from the source at relatively high speed,
gradually dying off as their energy is absorbed
by the medium.
What we call sound is simply the sensation
produced by the ear when stimulated by these
vibrations.
3
PROPERTIES OF SOUND A sound wave is a
series of pressure changes moving through the
air.
Amplitude(dB) is the difference between maximum
and minimum pressure defines the loudness
Frequency(Hz) is the rate at which the pressure
changes occur defines the pitch and timbre of
the sound
Wavelength(m) is the physical distance between
two maxima( or minima) depends on the speed of
sound in the medium and on the frequency
V l f Velocity
Wavelength Frequency
4
Sound Speed(m/s) Refers to the speed of travel
of the sound wave. This varies between mediums
and is also dependant on temperature.
In other materials, the speed of sound can vary
quite substantially.
5
AUDIBLE RANGE The ear can hear sounds
ranging from 20Hz to 20kHz.
It is most sensitive to frequencies between 500Hz
and 4000Hz, which corresponds almost exactly to
the speech band.
6
MEASURABLE CHARACTERISTICS Just how can we
measure a sound?
Acoustic Power (Watts)Measures energy output by
a source, that sound's ability to do work
Intensity (W/m²) The amount of sound energy
within a specific area normal to the direction of
propagation
Pressure (Pa)Measures fluctuations about the
local atmospheric pressure. Use of
root-mean-square (rms) rather that peak-to-peak
measures.
7
DECIBELS The units used to measure this ratio are
called bels. Two variables differ by one bel if
one is ten (1E1) times greater than the other,
and by two bels if one is one hundred (1E2) times
greater than the other. The bel is still a very
large unit and it is more convenient to divide it
into 10 parts - hence the decibel. The standard
threshold values given above corresponds to
exactly 0 decibels. The actual average threshold
of hearing at 1000 Hz is more like about 4
decibels, but zero decibels is a convenient
reference.
8
Sound pressure level (SPL) or sound level Lp is a
logarithmic measure of the rms pressure
(force/area) of a particular sound relative to a
reference sound source. It is usually measured
in decibel (dB(SPL), dBSPL, or dBSPL).
It can be useful to express sound pressure in
this way when dealing with hearing, as the
perceived loudness of a sound correlates roughly
logarithmically to its sound pressure
9
(No Transcript)
10
Sound Power refers to the absolute power of a
sound source (in Watts) whereas Sound Power Level
refers to the magnitude of that power relative
to a reference power (in dB).
The sound pressure ( dB) of a given speaker can
be easily calculated knowing the sensivity and
the driving power (W)
SPL Sensivity LW
11
Sound propagation in free field
12
Sound propagation in free field
Walking away from a sound source, the perceived
level of the sound decrease
This is known as the standard inverse square law
for point sources. Practically results in 6 dB
reduction in relative intensity per doubling of
distance.
NOTE 1 dB increase is barely audible
3 dB is a generally noticeable change
10 dB is considered as twice as loud
13
Mathematically looks like New level Old level
20xlog(old distance)- 20xlog(new distance)
LL20logD - 20logD
14
Sound Wind Temperature Gradients
15
Sound and Barriers a matter of wavelength
16
Room Acoustic
Sound waves will propagate away from the source
until they encounter one of the room's boundaries
where, in general, some of the energy will be
absorbed, some transmitted and the rest reflected
back into the room.
17
Part of the sound emitted from the source will go
directly to the listener, part will be absorbed,
and reflected by walls.
The indirect sound after several reflections from
different surfaces becomes diffuse creating a
steady state field
.
In the created field the sound does not have
directivity and the inverse square low doesnt
hold anymore.
This is called reverberation
18
When the sound source is turned off, direct sound
will stop and only the reverberant field will
remain
After some seconds even the reverberant field
decays.
The length of time taken for a sound to decay 60
dB after the source has ceased transmitting is
defined as Reverberation time
19
Room Acoustic depends on
Primary volume, shape, linear dimensions

• Volume defines if a sound reinforcement system
  is needed or not. Defines directly the
  reverberation time

• Shape flat, parallel walls, domes, defines
  echoes and reflections

These are fixed and can be hardly changed
Secondary Walls, Ceiling, Materials, Furniture
Treatment can be suggested to improve the room
acoustic
20
Different materials reflect sound in different
way
21
Marble
Carpet with foam base
22
Reverberation time, RT60, depends on room
dimensions and absorption of the walls
whereRT is the reverberation time in seconds,V
is the volume of the room in cubic meters, is
the average absorption coefficient of the room,
andS is the total surface area of the room in
square meters
The reverberation time affects most of the
acoustic features of the room.
23
In acoustic, rooms with smaller reverberation
times are appropriate for speech, whereas spaces
designed for music require longer reverberation
times.
24
More complex equations was developed to take care
of different environment
   
25
In every room coexist a Direct Sound and a
Reverberant Field
There is a point in which the Direct SPL and
Reverberant SPL are equal. This point is at a
distance , from the source, called CRITICAL
DISTANCE
Where Q is the directivity of the source
Every point farther than the Dc from the source
will hear just the Reverberant field. The inverse
square law is no more valid
26
What is a good sound?
Fidelity. Is given by the frequency response. It
depends on each item of the audio chain
Loudness must be sufficient to achieve the
required effect. Is determined by the dynamic
range of the sound system
Intelligibility is linked by the signal/noise
ratio and the direct-to-reverberant sound ratio
at listeners ear.It depends directly on room
acoustic
Room Acoustic is as important as the sound system
itself.
27
Audibility ? Clarity
ability to detect the structure of a sound
ability to hear a sound
This distinction is more important in speech than
in music
28
Speech

• Is made of consonant and vowels
• Vowels range from 250-500hz, carry power
• Consonants range from 1-4kHz, carry information

Lose consonants Lose intelligibility
29
Intelligibility
Is not a physical quantity as Ampere, Volt, Watt
Measure of the degree of understanding spoken
language
There are many index to express this degree, many
way to measure, and predict it
30
Factors Affecting Intelligibility
In on-to-one conversation there arent any
problems of intelligibility
Sound System Bandwidth and Frequency Response
Room Reverberation
Geometric Factors
Signal-to-Noise Ratio
Distortions
Non Linear Factors
31
Bandwidth and Frequency Response
Sound system have to guarantee a response from
100 to 10000 Hz.
Limits are fixed by worse performance
32
Signal to Noise Ratio
SPL must be adequate and heard comfortably
(normal conversation ?70-90 dB)
Noise masks direct sound and lowers
intelligibility
Increasing S/N ratio increases intelligibility
Intelligibility becomes independent from S/N for
S/Ngt25 dB
33
Reverberation and Reflections
Long RT60s decrease intelligibility
Late reflections (gt 50 ms) smear and blur direct
speech
Early reflections ( lt 35-50 ms) are perceived as
reinforce
34
Distortion
Clipping
Are form of NOISE
Intermodulation
Acoustic distortion
Specification of various items that compose the
sound system have to be carefully studied
35
Measure and Predicting Intelligibility
Design for speech intelligibility is as important
as design for gain, SPL and coverage
While it is quite easy to calculate SPL and RT60
there arent models to calculate Intelligibility
degree taking care of all parameters
There are more way and several index to express
Intelligibility Degree Subject Based ( AI,
ALCONS) Quantitative ( STI, RaSTI)
36
Predicting ALCONS
ALCONS is an index expressing Intelligibility
degree, in terms of lost consonants in the
talker-listener path
The simplest Peutz formula take care of
Directivity, RT60, Room Volume, Number of
Speakers, Distance Loudspeaker-Listener
The modified Peutz formula includes also Direct
SPL, Reverberant SPL, and Noise SPL
37
ALCONS INDEX
High Qs and Large Vs improve ALCONS
Long Ds, long RT60s lowers ALCONS
This formula fails when strong non-linear effect
are present
38
STI and RaSTI
These methods are fully independent of human
being and are fully quantitative
Take care of all factors affecting the
intelligibility because measures the corruption
of a speech based signal during the
talker-to-listener path
Varies from 0 no intelligibility to
1 perfect intelligibility
39
STI and RaSTI main features
Replace speech with a high frequency noise
(consonants-vowels) modulated in amplitude by a
low frequency signal (phonems)
Knowing the m(f) means predict intelligibility
40
Every reduction to the amplitude of the
modulation function m(f) results in a degradation
of intelligibility degree
41
Alcons and RaSTI are linked
42
Common Intelligibility Scale (CIS)
There is a common scale to simplify to define the
limits of acceptable intelligibility
CIS0.7
Standard CEI EN60849 states that CISgt 0.7
ALcons12 STI0.5
43
Speech Intelligibility Optimisation Practical
Criteria
Sound quality and intelligibility are not the
same thing

• Aim the loudspeaker to the listener keep as
  much sound as possible off the walls and the
  ceiling
• Provide a direct line between loudspeaker and
  listener
• Ensure adequate bandwidth
• Avoid frequency response anomalies (corner bass
  increment)
• Minimize D where possible
• Ensure S/N ratiogt10dB
• Avoid delaysgt 50ms ( inter speaker spacinglt 15m)
• Use high Q in reverberant environment
• Minimize SPL variations
• Improve RT and acoustic environment

44
Sound System is the complex of electromechanical
items that ensure the reproduction, amplification
and diffusion of speech, sound and messages
coming form a live source or a recorded one
Speakers Line
45

• Electro-Acoustics 
• Reasons for using sound amplification equipment
  within an architectural context
• To increase the sound level when a sound source
  is too weak to be heard.
• To provide additional sound to audiences beyond
  the intended range of the source.
• To project sound back to the stage for the
  benefit of the performers.
• To alter the Reverberation Time or other
  impression of an auditoria.
• To reduce the relative effects of background
  noise.
• To provide paging, information or warning
  facilities.
• To reproduce electronic or recorded material.

46

• A sound system is basically composed of
• Electro-Acoustic components ( speakers,
  microphones detectors)
• Electronic items (mixer, amplifier, digital
  processors, music/message sources)
• Environment ( Room Acoustic, RT60)

Any result is a mesh of these components, and the
lower quality component will lower the
performance of all the others together
47
A Sound Reinforcement system is a system for
accurately amplifying, reproducing, and sometimes
recording audio, so that persons not near the
original source may experience the sound as if
they were.
PA system, controls to mix the signals coming
from the various microphones or other input
sources (such as Tuner, CD, MP3 and so on).
48
How to approach a study of a sound system
It is advisable to begin your study of a sound
system with the loudspeakers, after which the
amplifier power and model can be defined, and
finally the sound sources and appropriate
connection system can be selected.
Specifically, you need to
1) Establish the required system functions on the
basis of the users needs. 2) Analyse the
characteristics of the environment 3) Choose the
loudspeakers on the basis of the nature and
dimensions of the space, the type of message to
be transmitted (speech/music), and the noise
level of the environment. 4) Choose amplifiers
that are suitable for driving the speakers
selected and with a sufficient number of inputs
for all the sounds sources. 5) Define the sound
sources (microphones, tuners, cassette, players,
etc.). 6) Evaluate the connection system for the
speakers and establish cable sections.
49
The Sound System Design flow chart
50
Speaker Placement There are essentially two types
of speaker system
A distributed system/multi point diffusion
A centrally located system
51
Centrally Located System
Minimize the Reverberant field but can result in
long speaker/listener distances
Need for Loudness Calculation
Need for Coverage Calculation
52
9 x H1315 Central Cluster
53
Distributed system
Increase the Reverberant field, lower the
speaker/listener distance

54
MQ60H 1800 coverage, 97dB 3m away
Small-medium size spaces
DM61 1200 coverage, 96dB 3m away IP55
55
Medium-Large size spaces
56
Speaker positioning in corridors
MQ30P 92dB after 10 m
57
(No Transcript)
58
Several RAIN DIFFUSION coverage
59

• A rule of thumb for distributed system is
  calculate the ratio

60
SPL 1W/1m
90dB
66dB
78dB
84dB
72dB
1m 2m 4m 8m
16m
1W
93dB
69dB
81dB
87dB
75dB
1m 2m 4m 8m
16m
2W
96dB
72dB
84dB
90dB
78dB
1m 2m 4m 8m
16m
4W
61
Line Loss and Wire Section
The load is considered to be concentrated at the
end of the line. If the speakers are distributed
along the line, the section can be almost
halved.
62
How to connect speakers when

• lot of speakers connected,
• long speaker lines,

Constrains of a traditional low impedance
connection
Series little power delivered to speakers
Series/Parallel reasonable impedance but complex
Parallel Hi currents!!!
Heavy loss due to wiring resistance
63
100V Line connection.
Thanks to a step-up transformer and step-down
transformers connected to speakers the problem
related to impedance matching is solved
High transformer impedance and high line voltage
minimize the wiring power loss
Failure of a speaker do not affect the line and
the other speakers
64
RCF Case Studies

• Abu Dhabi Airport
• Rome Exhibition Centre
• Big Stand for Parade in Singapore
• Auditorium in Singapore

65
RCF RX3000 in Abu Dhabi airport

• Programmable total acces matrix for sound system
  management and control ( max 64 IN x 128 OUT)
• 2 Digital paging consoles with 16 programmable
  Keys
• 7 Analogue Consoles with 4 programmable keys
• 1 five CD charger for BGM
• 1 Automatic Announcement System
• 14 output, with monitored amplifiers/lines

66
(No Transcript)
67
(No Transcript)
68
(No Transcript)
69
(No Transcript)
70
RX3000 full access matrix
71
(No Transcript)
72
(No Transcript)
73
(No Transcript)
74
RCF System in Rome Exhibition Centre

• 22 big and acustically difficult buildings
• Each pavillion must have its own
  evacuation/paging/bgm system
• Each building have to be digitally connected (
  Ethernet\optic fibre) to a central control room
• IP control requested

75
Study of the Room Acoustic and desing of the
sound system
The main building is 144 x 72 m with a concave
ceiling of 8-16 m high ( about 125000m3) with a
measured RT60 of 3.93 s
The building is too big for a centrally located
sound system and too reverberant and high for a
high distributed system ( rain diffusion). There
are some architectural constrains to place the
speakers
The idea is to use clusters to cover uniformelly
part of the area and minimize the Reverberation
Field
76
(No Transcript)
77
One of the siutable solution is use 10 round
cluster of 8 speaker each.
The coverage is quite good, but the cluster is
complex
78
Try another kind of cluster 2 array of 4
ART110Ti, side by side
After some fair calculation, the decision to
build and measure this cluster
79
The coverage is good, better than the previous
solution!
80
Real measurment was better than the expected from
the calculation
This solution was accepted!
81
The suggested placement of the speakers in the
room
82
The real placement of the speakers according to
the architectural constrains
The cluster is perfectly integrated in the
structure
83
(No Transcript)
84
(No Transcript)
85
(No Transcript)
86
RCF in very large outdoor stands

• 30000 seat stands 240m lenght
• 150x100 m stage
• Strong constrains to the speakers installation (
  masts on stands sides)
• Limitated noise pollution behind the stands
• IP55 protection required ( near the sea)

87
An overview of the area The stand The parade
stage
88
The costumer asked for CS6940 columns
The SPL is too low
89
The best solution is to use horns
H6045 Long-throw horn
HD410/T
HD210/T
90
Speakers aiming was carefully studied to cover
the area in unifom way and avoid acoustic
pollution
Dealyed speakers lines was employed
91
If we consider that crowd noise can be more
than 80-85dB, we need 20 dB S/N
The coverage was really uniform and always
greater than 100 dB
92
The noise level on the walls, behind the seating
area varies from 80 to 75 dB, the same level of
normal traffic noise pollution.
93
RCF in Auditoria
The system is for Music/Speech in a an auditorium
The costumer required a stereo system
The costumer need the study of the room acoustic
94
Starting from the drawings, a model is built
95
Agreed with the architects, several materials
were considered ( carpet, gypsum boards, wood,
chairs)
The RT60 for this environment was calculated and
optimized 1.7 sec at 1KHz for empty room
96
High quality speakers were chosen from
If it is considered a 60dB of background noise,
for this auditorium is request a minimum of 85
dB ( 25 dB of signal to noise ratio usually
required) with 20 dB headroom.
ACUSTICA C5212
97
A uniform coverage was obtained over the whole
listening areas
The Direct SPL at 1KHz is about 108dB 4 dB The
goal has been reached
98
RaSTI were calculated with long Peutz formula at
2KHz to evaluate the intellegibility of speech in
the room
RaSTI ranges from 0.50 to 0.57. The goal has
been reached!!
99
Thank you for your kind attention!
Dear listener, please find more suggestion in
RCF Sound System Design and Installation Guide
The Rules of Sound