1.Acoustics Seminar_Reading.docx, Page 1 of 33.

The Sonic Arts Center

@
The City College of New York

Acoustics Seminar
Music 21700

Paul Kozel
Edits and addendums, Jonathan Perl

1.Acoustics Seminar_Reading.docx, Page 2 of 33.

The Sonic Arts Center @ The City College of New York

All Fig. numbers are examples from An Introduction to the Creation of Electroacoustic Music, by Samuel Pellman
(out of print).
What is Sound?
Sound starts with the movement of an elastic medium (i.e. a string). The movement of this medium
(sound propagation) disturbs the surrounding air molecules, which creates regular and/or irregular patterns of
compression and rarefaction of these air molecules. Air molecules move toward each other during
compression and away from each other during rarefaction. High air pressure is created during compression while
low air pressure is created during rarefaction. These compressions and rarefactions are called vibrations.
Vibrations travel to the ear where they are processed into perceptions of either noise or pitched sound or
combinations of the two.
Noise is created by irregular patterns of compression and rarefaction while pitched sound or frequency
is created by regular patterns of compression and rarefaction. These vibrations radiate in various directional
patterns at a constant speed of 1,130 feet per second (the speed of sound). The speed of sound is dependent on
the medium and the temperature of the medium. Sound will travel through 72 Fahrenheit dry air at 1,130
feet-per-second.
Irregular vibration = Noise

Regular vibration = Pitched sound/frequency

(See last page of this handout for more information on noise.)

1.Acoustics Seminar_Reading.docx, Page 3 of 33.

The Inner Ear: Pitch Perception and Frequency Masking

The Cochlea is the part of the inner ear containing several membranes, including the basilar membrane.
Sounds excite the fluid within the inner ear via the action of the stapes which are in contact with the oval window.
On the surface of the basilar membrane lies the Organ of Corti, which contains between 20,000 and 30,000
reedlike fibers/hairs, along its length, and these change properties: they are short and stiff at the start of the
basilar membrane, and become increasingly longer and less stiff further down the membrane.
Because of their variable stiffness and length, the hairs have different resonant frequencies. Nearer to the
middle ear, the shorter stiffer hairs will be excited by higher-frequency components of the sound. Further on
down the membrane, the longer looser hairs will be excited by lower-frequency components of the sound. The
basilar membrane fires electrical impulses based on these excitations. These impulses travel via the cochlear
nerve, which sends them on to the cerebral cortex of our brains. Our brains, in turn, have the amazingly
complex job of interpreting and making sense of the impulses arriving from the basilar membrane. Among the
most crucial interpretations is that of pitch: we can perceive specific identifiable pitches based on the location of
firing of nerve impulses from the basilar membrane.
Because most pitched sounds are not sine waves but
instead contain many sine waves at different frequencies simultaneously, the concurrent firing from multiple
locations can not only result in a perception of pitch, but in a perception of tone or timbre.
If two fundamental tones are very close in pitch, they will excite areas that are very close to each other on the
basilar membrane. In this case, our brains may be unable to parse the two distinct pitches. Whichever tone is
louder and is causing the greater amplitude of vibration will be perceived as a distinct pitch, and the adjacent pitch
will be masked, only contributing to an increase in perceived volume.
Sounds with greater sound pressure amplitude will cause the hairs of the membrane to vibrate with a greater
range of motion/displacement. The resulting nerve impulses are interpreted by the brain as being of sounds with
a greater perceived loudness.

1.Acoustics Seminar_Reading.docx, Page 4 of 33.

The Mechanics of a Vibrating System The Damped String

In order to demonstrate a vibrating system for pitched sound we will employ a plucked, vibrating string as
demonstrated in Fig. 1.1 below. When a string is plucked it is displaced (Fig. 1.1a) or moved away from the
point of rest/equilibrium. The point of equilibrium or rest is called the node. The point of maximum displacement
from the node is called the antinode. The antinode also represents extremes in air pressure. When on the upper
side of the node, antinodes represent maximum compression. While on the lower side of the node, antinodes
represent maximum rarefaction. The antinode represents the amplitude of the sound. The further the antinode is
displaced from the node the louder the sound (Fig. 1.6).
When the string is released from the point of displacement, restoring force begins to act on the string,
which attempts to bring the string back to the node or point of rest (Fig. 1.1b). However, momentum pushes the
string past the node (Fig. 1.1c) to a point that is nearly the same distance from the resting point as the initial point
of displacement. This point however, is on the opposite side of the node and represents another antinode. The
string then begins the same journey, now in the opposite direction, moving from Fig. 1.1d back to Fig. 1.1f.
During the strings movement, friction from the surrounding air molecules creates drag on the string so that the
string never returns to the point of the original displacement and never is able to reach the equivalent point on the
opposite side of the node. The force of friction will eventually cause the string to stop vibrating altogether
(damped string) unless more energy is supplied by displacing the string again in some manner. The gradual
dampening of the motion of the string is called decay.

Upper Antinodes = Points of Maximum Compression

1.Acoustics Seminar_Reading.docx, Page 5 of 33.

Lower Antinodes = Points of Maximum Rarefaction

Pitched Sound/Frequency and Pitch
The movement from Fig 1.1a through Fig. 1.1f is called a cycle. A cycle consists of one compression
and one rarefaction (Fig. 1.4). Frequency is the number of cycles occurring during the time period of one second
(440 cycles per second, etc.). Cycles per second are also called Hertz after the 19th-century German physicist
Heinrich Hertz. Pitch is a letter name applied to certain frequencies (A - 440 Hertz etc.). The higher the
frequency the more cycles take place in one second.
The human ear has the ability to perceive the number of regular vibrations (cycles) that take place in one
second. In short, it can distinguish between a regular vibration of 500 times a second, 200 times a second, etc.
The range of frequencies most humans perceive is approximately 20-20,000 cycles per second.
Frequencies above this range are called ultrasonic. Frequencies below this range are called infrasonic
or subsonic.

Frequency of a String
The frequency a string produces is dependent on three factors: string length, the tension of the string,
and the mass of the string. Longer strings produce lower pitches. Increased tension increases frequency.

1.Acoustics Seminar_Reading.docx, Page 6 of 33.

Increased mass decreases frequency. All strings of a guitar are of the same length. Differences in frequency
between the strings of a guitar are not dependent on length but on tension and mass. The tuning pegs adjust the
tension and the different string thickness combine to create six distinct frequencies. Note that some strings are
made of a solid core (top strings) while others are solid core covered with over-wound strings (bottom strings).
The over-wound string increases the mass and therefore lowers the pitch of the string. The piano uses a
combination of differences in string length, tension, and mass to produce different frequencies.
The longer the string The lower the frequency
The more tension on a string The higher the frequency
The more mass of a string The lower the pitch

1.Acoustics Seminar_Reading.docx, Page 7 of 33.

Physical Properties of a Wave

Since each frequency radiates a specific number of cycles over the time of one second at the speed of
sound, each cycle will have specific physical length called a wavelength. Wavelength is commonly designated
with the Greek letter lambda ().
Low frequencies have long wavelengths (since there are less cycles per second) while high frequencies
have shorter wavelengths (since there are more cycles per second). The formulas below demonstrate the
relationship between wavelength, frequency, and the speed of sound.

F (frequency in Hertz) x W (wavelength in feet) = S (speed of sound)

Therefore:
565 Hertz x 2 feet = 1,130 feet per second
Conversely:
S/W=F
1,130 feet per second / 2 feet = 565 Hertz
and
S/F=W
1,130 feet per second / 565 Hertz = 2 feet
Phase Angle and Phase Shift
Phase Angle represents a specific point in time during the cycle of a wave. The length of time it takes to
complete a single cycle is called a period. The passage of time during the period is divided into 360 equal parts,
or degrees, and can be plotted about a circle (see drawing below).
Phase Shift indicates the amount of time that exists between two identical waves. The time difference
between these identical waves is classified by where the first wave is in its cycle when the second wave begins.
For example, if the second wave begins when the first wave has already reached 90 of its cycle the waves are
said to be phase shifted by 90 (see drawing below). If identical waves begin at the same time they are said to be
in phase with each other. However, if these waves begin at different times they are out of phase with each
other.

1.Acoustics Seminar_Reading.docx, Page 8 of 33.

Phase Angles of a Single Wave

Phase Shift of Two Identical Waves

Dotted Waves are Phase Shifted from the Original Wave (Green)

1.Acoustics Seminar_Reading.docx, Page 9 of 33.

Constructive Interference
When two identical waves are in phase (starting at the same time) the amplitudes of the these waves are
summed. This is called constructive interference or phase summation (Fig. 1.10).

1.Acoustics Seminar_Reading.docx, Page 10 of 33.

Deconstructive Interference
If two identical waves are 180 out of phase they will cancel each other out completely since they have
opposite yet equal energy. No sound will be perceived by the listener. This is called destructive interference,
or phase cancellation. (Fig. 1.11) If these two waves are 180 out of phase but have different amplitudes only
partial phase cancellation will occur. Partial phase cancellation will reduce the amplitude of the two waves based
on the differences in amplitude. The closer the two waves are in amplitude the greater the decrease in amplitude.

Polarity
Polarity simply refers to energy relationships between two identical waves. It does not refer to time
relationships. In the example below there is no time difference between the two identical waves. They both start
at the exact same time. However, there is a polarity difference since these waves are of equal but opposite
electrical energy. If these were acoustical waves they would be of equal but opposite air pressure. In the
example below the waves are said to have opposite polarities or polarity reversed. Two identical sine waves that
have opposite polarity or are phase shifted 180create the same result: zero output. However, they create this
effect by different means. (180 phase-shifting of waveforms that are more complex then sine waves - or of actual
audio tracks - does not result in total cancellation unless the waveforms are symmetrical - e.g., triangle waves and
square waves: see below for the complete discussion of phase-shifting effects, and also the section on geometric
waveforms.)

1.Acoustics Seminar_Reading.docx, Page 11 of 33.

Polarity Reversal
Phase Shift and Frequency
The amount of phase shift that results from a specific, fixed, time delay varies with frequency. In the
recording studio a digital delay processor has the ability to delay any frequencies that are sent to it by a specific
time value usually given in milliseconds. The digital delay processor then combines the original non-delayed
signal with the delayed copy to create various effects such as echo, phasing, flanging, and chorusing.
In the example below, if the four waveforms were delayed 5 milliseconds each signal would be phase
shifted by different amounts due to the fact that their periods are of varying length. So when these delayed
signals are combined with the non-delayed signals each frequency will have a different phase shift. In the
example below you can see that a 5 millisecond delay applied to a 400 Hertz waveform will create a 720 phase
shift (2 cycles or 360 x 2) while a 200 Hertz wave will be phase shifted 360 (1 cycle).
Dry/Unprocessed

Signal

(In a hardware processor this is an analog signal)

Delayed/Processed
Input

Signal

A/D

Signal

Convertor

Required for processors with

analog inputs

Mix/Output
D/A

(Delayed signal held in Shift Register)

Static delay time

Set with software fields, Midi

beat clock, switches, or LFO

Clock

Convertor

Required for processors with

analog outputs

Input

(Determines the lenght of delay)

Diagram of a Digital Delay Processor

1.Acoustics Seminar_Reading.docx, Page 12 of 33.

Phase Shift and Perception

Phase shift created by delay times over 60 milliseconds or so are not really considered phase shift since
the time period between these two waves is discernable by the human ear. Once the ear can identify the time
period between the original signal and the phased shifted/delayed copy it is considered a type of echo effect, not
phase shift.
If the copy of the original signal is delayed by less than 30-40 milliseconds and its mixed with the original
signal you will not actually be able to hear the time difference between the two signals. The time differences will
not be perceived as being distinct or individual. Instead they will be perceived as a collective whole much like the
eye perceives a single image of a spinning coin consisting of both heads and tails. In both cases the eye and ear
have passed their threshold to hear and see distinct reiterations.
When a signal is delayed less then 30-40 milliseconds and is combined with the original signal the
resultant shift in phase of the various frequencies will create patterns of constructive and deconstructive
interference between the two waves. These patterns create what is known as the comb-filter response (see
diagram below). These patterns will vary depending on the length of the delay. When the time differences

1.Acoustics Seminar_Reading.docx, Page 13 of 33.

between two identical waves are less than 30 40 milliseconds listeners are actually detecting changes in the
comb-filter patterns!

Comb Filter Response

Delay Times and Listening Responses

Below 40 ms.

40 - 60 ms.

60 - 100 ms.

Diferences in phase. Warbling/doubling effect.

reiterations.
Very
Comb filter response. Blurred

Flutter.
rapid reiterations.

Over 100 ms.

Echo

Phase Shifting with Microphones and Speakers

Phase shifting can be inadvertently created in the studio when several microphones pick up the same
wave at different distances from the source of the wave (see Phase Problems with Microphones below). Each
microphone will pick up the wave at a different place in the cycle since they are at two different distances from the
sound source. These arrival time differences will create phase differences when the waves are combined
together. Some of the differences in phase can create an audible reduction in the amplitude of some of the
frequency components in the output of the two microphones when they are combined together (deconstructive
interference).
Changing the distance of one of the microphones from the sound source will change the phase
relationship between the two microphones and may help to reduce the amount of deconstructive interference or
comb-filter effect. Also many mixing consoles and outboard microphone preamplifiers will have a phase switch
that can be engaged on one of the microphone inputs to change the phase by 180. Many times this can help
restore amplitude lost due to phase cancellation.
This phase switch is a real misnomer. It really should be called a polarity switch since it simply
changes the electrical properties of the signal through polarity reversal, not through any time delay.
The ear is extremely sensitive to phase shift. Even phase shifting of less than 1 ms are perceivable! In
fact, the ear can perceive time differences that are as small as .030 microseconds (.03s)!!!!
Two speakers that are used to create a stereo image are very phase dependent. The sound from each
speaker must reach the listeners ears in-phase in order for the listener to perceive the sounds accurately. Phase

1.Acoustics Seminar_Reading.docx, Page 14 of 33.

differences can affect the frequency response, spatial placement, and depth perception of the sounds emanating
from a pair of speakers. Sitting in the sweet spot between two speakers is simply being equidistant from each of
the speakers so that they will produce the most accurate picture of the audio.

Phase Problems With Microphones

30' From Source

1 Wavelength
10'

1 Wavelength
10'

1 Wavelength
10'

90

Microphone Microphone
B
A

110 Hz Wave
Sound Source

180
270

25' From Source

Mics pick up the same wave 180 out of phase.
Note that this is one sigle frequency
.

30' From Source

1/2
1 Wavelength
20'

Wavelength
10'
Microphone
A

90
Microphone
B

55 Hz Wave
Sound Source

180

25' From Source

Mics pick up the same wave 90 out of phase.
Note that this is one sigle frequency
.

Heterodyning (Amplitude Modulation)

When two waves are played together that have the same amplitude and are close to each other in
frequency (less than 1 to approximately 20 Hertz) the two frequencies, over a period of time, will sum their
amplitudes when they are in phase and cancel their amplitudes when they are out of phase. The number of
amplitude summations or "beats" that occur during one second can be calculated by subtracting the lowest
frequency from the highest. For example, if you have a frequency of 500 hertz and 505 hertz you will hear 5
beats occurring each second (Fig. 1.12). Heterodyning is used to thicken up a sound to make it more interesting
to listen to. For instance, the mid to upper range notes on a piano actually use three strings for each note. Each
of the three strings is tuned slightly differently to create the Heterodyning effect. Once the difference in frequency
reaches the 15-20 Hz range the listener at some point will begin to discern two distinct frequencies without
amplitude fluctuations instead of one frequency with amplitude fluctuations.

1.Acoustics Seminar_Reading.docx, Page 15 of 33.

1.Acoustics Seminar_Reading.docx, Page 16 of 33.

Modes Of Vibration
Inversely Proportional Law- The frequency (number of cycles per second) of a vibrating string is inversely
proportional to its length. In short, as frequency increases, the length of string decreases. Conversely, as the
frequency decreases, the length of the string increases. If you play a single guitar string, you can produce
progressively higher frequencies by shortening the vibrating length of the string (see diagram below).
vibrating length

Neck

whole string vibrating

Bridge

vibrating length

Neck

Bridge

1/2 string vibrating

vibrating length

Neck

Bridge

1/3 string vibrating

The ratio that exists between any two frequencies is called an interval. The interval created when the
string vibrates as a whole (open string) and when it vibrates at half its length has a ratio of 1:2. Therefore, if the
open string is 440 Hertz then the string vibrating at half its length is 880 Hertz or 440 x 2. This interval is called
the octave. Each doubling in frequency produces an octave interval. It represents a fixed ratio, not a specific
number of Hertz between frequencies (for example all octaves being 100 Hertz apart). Therefore, its logical to
deduce that the intervals express a logarithmic relationship.

1.Acoustics Seminar_Reading.docx, Page 17 of 33.

The graph below shows the relationship between the open string (vibrating as a whole) and other
vibrating modes 1/2, 1/3, 1/4 etc. expressed in ratio, interval name, and frequency. These ratios and their
resultant intervals are called pure intervals because they exist in natural vibrating systems like string and wind
instruments.

string length:whole 1/2

frequency:

67
fx1

1/3

134
fx2

ratio:

1:2

interval:

octave

1/4

201
fx3

2:3

268
fx4

3:4

fifth

1/5
335
fx5

4:5

fourth

maj. 3rd

The lower number of a ratio expresses the lower frequency while the higher number the higher frequency.
The frequencies of 67 Hertz and 134 Hertz (2x67) create the ratio of 1:2. 134 Hertz (2x67) and 201 Hertz (3x67)
create the ratio of 2:3. 201 Hertz (3x67) and 268 Hertz (4x67) create the ratio of 3:4, etc. The interval names
listed above represent some of the consonant intervals. These intervals have simple ratios and are considered
pleasant or neutral sounding. Dissonant intervals are expressed with more complex ratios and are considered
more harsh sounding than consonant Intervals. The diminished fifth is a dissonant interval due to its sound and
the ratio of 45:32.
The formulas below demonstrate how to determine the second frequency of an interval when one is provided.
Interval
Perfect 5
Perfect 5
Perfect 4
Major 3rd

Ratio
3:2
3:2
4:3
5:4

Up or Down
Up (2:3)
Down (3:2)
Up (3:4)
Down (5:4)

Starting Pitch
A-440
A-440
A-440
A-440

Formula
440x32=660 Hz
440x23=293.3 Hz
440x43=586.6 Hz
440x45=352 Hz

Other Pure Interval/Ratios

Minor 2nd
25/24
Major 2nd
9:8
Minor 3rd
6:5
Diminished 5th
45:32
Minor 6th
8/5
Major 6th
5:3
Minor 7th
9:5
th
Major 7
15:8

Tuning Systems - Pure and Tempered Intervals/Ratios

When pure or just ratios are used to create a tuning system for a musical instrument (how each note of a
scale is tuned) each key is tuned to the tonic point of reference. To tune an instrument to the C Major scale for
example, you would start with the tonic frequency (C) and then use the pure/just ratios to locate the frequencies of
the remaining pitches. Since each scale has its own unique point of reference this type of tuning doesnt work
well for music that changes keys. While each scale may sound in tune with itself its not necessarily in tune with
other keys since they use different points of reference.

1.Acoustics Seminar_Reading.docx, Page 18 of 33.

An equal tempered tuning system divides all notes of the chromatic scale into twelve equal parts.
Intervals are not derived by pure or just ratios but by using a constant frequency multiple between notes of a
chromatic scale (1.059463). This makes all intervals and all scales equal. Since all half steps are equal in this
system we further divide the equal half step into Cents. There are 100 cents in every half step so 50 cents
would equal a quarter step.
Overtone Series
All strings have the ability to vibrate in many different modes simultaneously. Strings can vibrate as a
whole (a below), in halves (b below), in thirds (c below), etc. all at the same time. The entire string is moving in
each mode but various parts of the string may be moving in different directions as demonstrated below.
vibrating length

a . Neck

Bridge
whole string vibrating
vibrating length

b . Neck

Bridge
vibrating in halves
vibrating length

c . Neck

Bridge
vibrating in thirds

A string vibrating in two equal halves (a below) will produce the same frequency that half of the string produces (b
below). A string vibrating in thirds will produce the same frequency that 1/3 of the string produces, etc.

vibrating length

a . Neck

vibrating length

Bridge

b . Neck

Bridge

vibrating in halves

(440 Hertz open string. 880 Hertz

when string vibrates in two halves)

1/2 string vibrating

(440 Hertz open string. 880 Hertz when

only half the string vibrates)
String Length is related to frequency inversely: Therefore, F = Frequency, L = string Length.
F x 2 = L 2 and L x 2 = F 2 etc.
Each whole # multiple of the fundamental frequency represents the next pitch, or mode of vibration, in the
harmonic overtone series (see below)

1.Acoustics Seminar_Reading.docx, Page 19 of 33.

The Overtone System

While it is possible for a string to vibrate in just one of these modes at a time to create a simple vibration it
usually creates a complex vibration resulting from the simultaneous vibration of many modes simultaneously.
When a guitar string is plucked the string is vibrating as a whole, in halves, in thirds, fourths, etc. simultaneously!
These different modes of vibration are called partials. The first partial or mode of vibration (called the
fundamental) generally produces the loudest amplitude and therefore is the pitch we discern. However, even if
the fundamental is not the loudest partial you will still perceive the fundamental pitch because of the consistent
pattern of partials above the fundamental. In addition, when you calculate the difference between any two
adjacent partials they will always give you the fundamental frequency! In short, since the ear recognizes the
pattern of partials above the fundamental it perceives the fundamental. In fact it can perceive fundamentals that
are completely missing! Partials above the fundamental are called overtones or harmonics since they exist
over the fundamental.
This simultaneous vibration of multiple modes creates the harmonic overtone series or just overtone
series that exists naturally in nearly all periodic vibrating systems that deal with pitched sound. Although this
system represents an unchanging set of intervallic relationships, not all the partials have to be present, and the
amplitude relationships between the partials are infinitely variable. The inexhaustible variety of these two
variables produces the many different timbres or colors of pitched sounds.

The Harmonic Overtone Series

Ratios:

1:2

Interval:

3:4

Fifth

Fourth

Octave

Frequency in Hertz: 67
Fund

Partial #:

2:3

Overtone/Harmonic:
Mode of Vibration: Whole
String Lengths: 1 yrd

4:5

5:6

Maj 3rd

134

Min 3rd

7:8

*Min 3rd? **Maj 2nd?

268

335

402

*469

536

fx2

fx3

fx4

fx5

fx6

fx7

fx8

Fourths

Fifths

Sixths

1/4 yrd

1/5 yrd

Halves

1/2 yrd

201

6:7

Thirds

1/3 yrd

1/6 yrd

Sevenths

1/7 yrd

Eights

1/8 yrd

*The interval between the 6th and 7th partial (created by the ratio 6:7) does not yeild a true
B flat in the Equal Tempered System of tuning. A true tempered B flat would be created by a
ratio of 5:6 which would yield a frequency of 482.4 hz. The interval created by the ratio of 6:7
exists as a pure interval of the overtone series but does not exit in the tempered tuning system we
employ.
**Since the pure ratio of 6:7 does not create a tempered B flat than the interval from the 7th
partial to the 8th is not an interval of a major 2nd in the tempered system. The interval created
by the ratio of 7:8 exists only as a pure interval of the overtone series. A tempered major second
has a ratio of 8:9.

1.Acoustics Seminar_Reading.docx, Page 20 of 33.

Geometric Waveforms
The different combinations of present partials and their associated amplitudes create various waveforms.
The names of some of the standard geometric waveforms refer of the shapes they create when viewed with an
oscilloscope. Geometric waveforms use static or fixed amplitudes for all of the partials.
Sine Wave
The sine wave has only a single partial. No overtones exist in a sine wave.

Sine Wave

Triangle Wave

A = Amplitude
n = partial number

1.Acoustics Seminar_Reading.docx, Page 21 of 33.

Square Wave

Sawtooth Wave

http://www.audiocheck.net/audiofrequencysignalgenerator_sinetone.php

1.Acoustics Seminar_Reading.docx, Page 22 of 33.

Harmonic/Inharmonic Spectra
If all of the partials in a given waveform are whole-number multiples of the fundamental, the wave is said
to have harmonic spectra. (Fig. 7.2) Conversely, if some of the partials have a fractional relationship to the
fundamental the wave has inharmonic spectra (Fig. 7.3).

1.Acoustics Seminar_Reading.docx, Page 23 of 33.

Fourier Theorem, Analysis, and Synthesis

In the early 18th century Joseph Sauveau discovered the overtone series. In the early 19th century the
French mathematician Fourier developed a theory based on Sauveau's discovery. It stated that any complex
periodic vibration could be reduced to a summation of simple sine waves with specific amplitudes. This theory is
known as Fourier's Theorem. Analysis done on complex waves that break them down into simple sine waves
with intensity/amplitude envelopes is called Fourier Analysis or Fast Fourier Transfer Analysis (see below).

1.Acoustics Seminar_Reading.docx, Page 24 of 33.

Creating complex waves by adding together different individual sine waves is call Fourier synthesis (Fig. 7.5).

1.Acoustics Seminar_Reading.docx, Page 25 of 33.

Formant Regions
Sympathetic vibration or resonance occurs when a vibrating wave comes in contact with an elastic
medium and sets this medium in motion. Some vibrating spaces, due to the material of construction, size or
shape, will emphasize (add volume to) one or more frequencies due to sympathetic vibration. When these
frequencies are played the vibrating space will begin to sympathetically vibrate, causing an increase in amplitude
of those frequencies. These emphasized frequencies are called formant regions. (Fig. 7.7)

Formant regions can also occur due to the absorption coefficient (ability to absorb acoustical energy) of
materials in a vibrating space. All materials employed in a vibrating space (whether inside a violin or a concert
hall) will have a unique absorption coefficient rating. All ratings are done by frequency range (amount of
absorption in a particular frequency band). Some materials will absorb high frequencies while others are best
suited for low frequency absorption. The combination of all of the materials in a vibrating space and their unique
absorption capabilities will create an overall frequency response with characteristic formant regions.
Another cause of formant regions is the overall dimensions and layout of a vibrating space that creates favorable
or unfavorable interactions between vibrations in the space. Waves emanating from one part of the vibrating
space (incident wave) can reach a boundary and reflect back (reflected wave) in the direction of the incident
wave. In some instances the interaction of the incident and reflected waves will create varying degrees of
constructive interference that in turn creates a formant region. The interaction of the incident and reflected waves

1.Acoustics Seminar_Reading.docx, Page 26 of 33.

resulting in constructive interference is called a standing wave. Standing waves are created by parallel surfaces.
Standing waves can be reduced or eliminated by avoiding parallel surfaces.
Measuring Amplitude - Objective, Empirical Measurement - Power and Intensity
The amplitude of a sound is the direct result of the amount of energy used to create the sound. One
empirical/objective measurement of amplitude can be accomplished by simply determining the amount of
energy/work spent over the period of one second to create the sound. This is a measurement of power.
W = E/T
W = Power E = Energy (measured in joules) T = Time (in seconds)
Another way of creating an empirical measurement of amplitude is to determine the amount of
energy/work that is dissipated in a vibrating space over the period of one second. This is a measurement of
intensity.
I = W/S
I = Intensity W = Power (measured in watts) S = surface area (in square meters)
An analogy to these types of measurements can be seen by looking at the inherent power of a light bulb
(determined by its construction and the electrical energy applied to it) and its resultant intensity in a room. This
intensity relies on the power of the light bulb but also the color of the walls, the size of the space, etc. Therefore,
in order for two rooms to have the same intensity of light they must both be of the exact same construction and
use the exact same power source.
A piano will have inherent power characteristics (determined by its construction and the mechanical
energy applied by the performer) but the perceived intensity will be determined by the distance from the piano and
the construction of the room that the piano is in.
Intensity determined by room characteristics

Power Source
Room Boundaries

Measuring Amplitude Human Subjective Perception of Amplitude

The subjective perception of amplitude change by the ear is accomplished through the comparative
analysis of amplitude A vs. amplitude B. Simply stated: amplitude A is louder than amplitude B. Or, that
amplitude is softer than this amplitude. The ear does not perceive amplitude as an empirical or absolute value as
it does with pitch/frequency (A-440 Hertz, etc.). But like frequency the ear perceives changes in amplitude in a
logarithmic fashion. The ear quantifies amplitude as ratios of power. For instance, a doubling of power is
perceived as the same change in amplitude regardless if the sounds are soft or loud: the ear perceives the same
amount of amplitude change no matter if the intensity ratio is, 1:2, 2:4, or 5,000:10,000 because each ratio
represents a doubling of power or a 1:2 power ratio. A 1:10 power ratio creates the perception of one sound
being twice as loud as another. The subjective perception of amplitude is called loudness.

1.Acoustics Seminar_Reading.docx, Page 27 of 33.

The graph below demonstrates how incremental doubling of intensity/power is perceived as a progressive
increase in loudness in equal increments.
Logarithmic
in IntensityLinear
Loudness Perception
Perception
Logarithmic
ChangeChange
in Intensity
|
Linear Loudness

Incremental
Incremental
doubling of
doubling
i n t e nof
sity
intensity

1:2

2:4

4:8

8:16

16:32

32:64

Equal
Equal
increments
incrementsof
lou
dn ess
of
loudness

6
The
listener
perceives
a progressive
The
listener
perceives
a progressive
increase
increase in loudness in equal
in loudness
in
equal
increments
increments

Amplitude measurements for all audio signals (electrical or acoustic) reflect the subjective way in which
we perceive changes in amplitude. This system employs logarithmic power/intensity ratios to provide linear
changes in loudness. The increments of amplitude change are called bels named after Alexander Graham Bell.
The bel represents a 10:1 power/intensity ratio. A 10:1 power ratio results in a doubling of perceived loudness
(this figure is disputed in the literature: loudness perception is affected by many factors and is quite variable and
subjective - see below section on the equal loudness curves). Each bel can be subdivided into 10 decibels or
dB. The decibel represents the unit of measurement that corresponds roughly to the amount a sound must be
raised in level to be heard as louder. Each incremental increase in dB is perceived as an equal change in
loudness that is accomplished with a logarithmic change in intensity.

1.Acoustics Seminar_Reading.docx, Page 28 of 33.

SUBJECTIVE PERCEPTION/LOUDNESS
Linear (equal) loudness changes are
perceived between one adjacent bel or
decibel and another
5 bels

OBJECTIVE REALITY = INTENSITY/POWER

Logrithmic intensity/power is required to
produce linear (equal) loudness
relationships
100,000 times the intensity of the point of reference
100,000:1 power ratio

4 bels
10:1 power ratio
dB (decibels)

3 bels
Each bel is 10x the
intensity/power or twice
the loudness of the
previous bel

10,000 times the intensity of the point of reference

10,000: 1 power ratio

2 bels
10:1 power ratio

1,000 times the intensity of the point of reference

1,000:1 power ratio

1 bel
10:1 power ratio

Point of Reference

100 times the intensity of the point of reference

100:1 power ratio
10:1 power ratio
10 times the intensity of the point of reference
10:1 power ratio
10:1 power ratio

(Threshold of hearing)

The diagram above shows the logarithmic relationship from one bel to the next. Note that all
measurements are using the threshold of hearing as the point of reference. The point of reference is an
absolute, empirical measurement (in power or intensity values) that represents the energy required to create the
softest sound an average person can hear. This is the value that all other amplitudes are compared to.
The diagram below demonstrates the logarithmic ratios and their linear equivalents in the decibel. Note
that a doubling of intensity/power (2:1 ratio) will always result in an increase of 3 dB.

dB scales and the levels in your DAW:

We know that an increase of 3 dB results in a doubling of Power or Intensity.
However, it requires a 6 dB increase to double Voltages, or to double Sound Pressure Levels.
You can observe this when working with metering in Logic or Pro Tools, which typically reference a voltage ratio,
not a power or intensity ratio. If you duplicate a solod track, the combined output of both the original with the
duplicate will register as 6 dB greater on the summing meter, not 3 dB greater.
If you find this confusing, take comfort, you are not alone.

1.Acoustics Seminar_Reading.docx, Page 29 of 33.

Logarithmic

Ratios Decibels
1

(db)

Bels

2
1 : 2)
(Doubling Intensity = 3 db increase

(3.01)

3
4

1:3

(4.77)

1:4

(6.02)

1:5

(6.98)

1:6

(7.78)

1:7

(8.45)

1:8
1:9

1:10

10

(9.03)
(9.54)
(10.00)

The ear can perceive an incredibly wide range of amplitude change from the very threshold of hearing up
to approximately the intensity ratio of 10,000,000,000,000:1 that represents the threshold of pain. The graph
below demonstrates some typical sounds sources, their power/intensity ratio compared to the threshold of
hearing, log, and the equivalent bel/decibel.
Sound
Threshold of pain
Jet taking off from 500 ft. away
Loud band in a club
Power saw
Subway
Heavy highway traffic
Busy street traffic
Close normal conversation
Office background noise
Quiet Conversation
Waiting room in doctors office
Recording studio ambient level
Breathing
Threshold of hearing

Power/Intensity Ratio
10,000,000,000,000:1
1,000,000,000,000:1
100,000,000,000:1
10,000,000,000:1
1,000,000,000:1
100,000,000:1
10,000,000:1
1,000,000:1
100,000:1
10,000:1
1,000:1
100:1
10:1
0:0

Log
10/13
10/12
10/11
10/10
10/9
10/8
10/7
10/6
10/5
10/4
10/3
10/2
10/1
10/0

Bel/dB
13/130
12/120
11/110
10/100
9/90
8/80
7/70
6/60
5/50
4/40
3/30
2/20
1/10
0/0

Below are a few power ratio and dB equivalents.

1:10 ratio = 10 dB or 1 Bel = Twice the perceived loudness

1:2 ratio = 3 dB increase
2:1 (half power) ratio = 3 dB decrease

What is a decibel? (link)

1.Acoustics Seminar_Reading.docx, Page 30 of 33.

Psychoacoustics - Equal Loudness Curves

In the early 1930s Harvey Fletcher and Wilden Munson discovered that the perception of loudness is
related to both the amplitude and the frequency of the sound. The chart below, called an Equal Loudness or
Fletcher Munson Curve, represents the actual intensity of various frequencies required for listeners to hear all
frequencies with the same loudness (using a 1 k Hertz signal as the point of reference).
Its clear from these loudness curves that we do not perceive the amplitude of all frequencies equally.
Low and high range frequencies need to be boosted in level in order for them to be heard as loud as the
frequencies around 1 k Hertz. Its also clear from the curves that as overall intensity increases our perceptions of
amplitude/frequency flatten out. Phons, as seen in the curves, represent a subjective loudness level similar to dB.
For instance, compare the 20 phons curve to the 100 phons curve. In the 20 phons curve an increase of
30 dB is required at 100 Hertz in order to hear it equally as loud as a 1kHz tone. However, in the 100 phones
curve only a 1 dB increase is required at 100 Hertz.
Most people listen to recorded music on speakers around 80-85 dB so the curve in this area is an
important reference level for all mixing and mastering engineers. If a mix/mastering engineer listens most of the
time in the 110 dB range they will tend produce tracks that are light in bass and high frequencies when playback
is at the 85 dB standard. Conversely, if a mix/mastering engineer listens most of the time in the 20 dB range they
will tend to produce tracks that are heavy in bass and high frequencies when playback is at the 85 dB standard.
In short, our perception of bass and high frequencies varies greatly with the overall level of playback.
It should also be self evident that low frequency sounds, like basses and kick drums, will represent a
major portion of the overall intensity/power of most recordings since they require so much more intensity to
compete with mid-range frequencies.

Psychoacoustics Loudness and Overall Listening Level

When overall listening levels are low a greater change in level will be required for us to perceive a Just
Noticeable Change or JNC in amplitude as would be perceived at a higher overall listening level. For instance, a
1 dB change in volume may represent the JNC at 80 phons . However, at 20 phons it might require a 4 dB
change in volume to represent the JNC.

1.Acoustics Seminar_Reading.docx, Page 31 of 33.

Psychoacoustics Loudness and Duration

The duration of a sound will have a significant impact on our ability to perceive the loudness of a sound.
Sounds that last less than a few tenths of a second (generally under three tenths/30 ms) will sound less loud than
the same sound (same intensity) with a longer duration.
Psychoacoustics Loudness and Frequency
Changing the intensity of a frequency may also create subjective perceptions by the ear particularly if the
frequency in question is high. A high frequency sound played with a lot of intensity will seem higher in pitch to
most listeners when compared to the same frequency played at a lower intensity. Conversely, a low sound
played with a lot of intensity will seem lower in pitch to most listeners when compared to the same frequency at a
lower intensity. This phenomenon is particularly apparent with pure tones.
Amplitude/Intensity Envelopes
The shape of a sounds amplitude in time is called an amplitude or intensity envelope.

R = Rate
L = Level

L1
Amplitude

R2
R1

L2

R3

Time
Finger Down

Finger Up

As the diagram above indicates, these shapes when plotted on a graph consist of time values (rate) and
amplitude values (level). R1 represents how long it takes the sound to go from no sound to the first amplitude
peak (L1). R2 represents the time it takes to go from the first amplitude peak to (L1) to the second amplitude
peak (L2). Finally R3 represents the time it takes to go from L2 to no sound at all.
In order to create envelope shapes on a synthesizer the rate and level values are used to instruct an
amplifier how and when to open and close. If you want the sound above to continue to sustain as long as the
note is held down you need to tell the synthesizer that L2 is the sustain level. Now when you depress a key the
synthesizer runs through R1, L1, R2, and holds on L2 as long as your finger is down. When you finger is
released from the keyboard the synthesizer then runs through R3 and then the sound stops (see Fig. 1.8 on next
page for various amplitude/intensity envelopes).

1.Acoustics Seminar_Reading.docx, Page 32 of 33.

1.Acoustics Seminar_Reading.docx, Page 33 of 33.

Noise
Irregular or random vibrations create noise. Noise can be created on a synthesizer with a noise generator. The
noise generator produces all frequencies in the audible spectrum and assigns random amplitudes to these
frequencies. Noise generators are defined by color, the color representing the part of the frequency spectrum (as
with light) that is emphasized. White noise is defined as having equal energy per unit frequency. This means
that there is the same amount of energy between 500 Hz and 501 Hz as there is between 1500 Hz and 1501 Hz.
Equal energy per unit frequency means that the noise is spectrally flat. Its' response is the same for all
frequencies. Pink noise contains equal energy per octave. For example, there is the same amount of energy
between 50 and 100 Hz as is between 7,500 and 15,000 Hz. Therefore, there is more energy in the lower part of
the frequency spectrum than the upper part. Pink noise derives its name from the color pink that is at the low end
of the light spectrum.