Introduction to Audio Engineering: Key Concepts

What two things have to happen for there to be sound?
Motion and a Medium for the sound to travel through

Waves are made of...
compressions and rarefactions
Compression is the high pressure point
Rarefaction is the low pressure point

What is a Sine Wave?
-simplest kind of wave
-very rare in nature
-contains only one frequency

Frequency
how many cycles per second

Amplitude
strength of signal
how high the peaks (compression) and how low the troughs (rarefactions)

Range of Healthy Human Hearing
20 Hz to 20kHz

Wavelength
the physical length of a complete cycle

Sound can either Reflect or Diffract. Explain both
Reflect - sound bouncing off an object
Diffract- bending of a waveform around an object

How big should a barrier be to stop diffraction around it?
A barrier must be larger than half a wavelength to stop that sound from diffracting around it, otherwise it will be like the barrier is not there

Compare frequency to its wavelength and how it interacts with environment
higher frequencies - shorter wavelengths - bounce off of things
lower frequencies - longer wavelengths - diffract and go everywhere

Complex tone (or waveform)
two or more frequencies (sine waves)

side notes -
almost everything we hear is a complex waveform
sine waves are boring - we mostly deal with complex waves

Octave
a doubling of frequency

Fundamental Frequency
first note, loudest one you hear, lowest, the first note you play with the full string vibrating

Timbre (tam-burr)
characteristic sound of something
the result of the relative amplitude (relative strength) of these various frequencies (harmonics)

White Noise
equal energy at all frequencies
(white is quite bright, highest octave has most energy, hurts to listen to if it's actually white noise)

Pink Noise
equal energy per octave
(20 Hz to 40 Hz has just as much energy as 10 kHz to 20 kHz -more of a roar of a noise)

Reverse polarity
flip positive and negative aspects of a wave
(frequency and amplitude are the same)

Decibel
Pm/Pr or I sig/ I ref

Why do we use Decibel?
1. Range of hearing is very wide - 10 trillion to 1
2. Human Response to loudness is logarithmic/looks like what it sounds like

How to perceive the sound as twice as loud?
Have 10x the amount of sound
When you have any sound, you need 10x more of any sound for people to perceive it as 2x the sound

Doubling of Power
+3dB

Robinson/Dadson and Fletcher/Munson Curves and what they tell us
Equal loudness contours
Graphs that show you how much amplitude is necessary across a range of frequencies for it to be perceived as the same loudness

Your perception of sound depends highly on listening level - mixing at various levels of dB can help make sure it's same across all levels

perception of music changes depending on how loud you listen to it
mix quiet - sound good loud
mix loud - sound like crap quiet

Peak Meter - PPM
made of light leds - can follow peaks in music very quickly (presence) -

VU meter - Volume units needle
Averages out sound, not quick enough to do peaks, looks like what it sounds like - follow the longer lasting amplitude

Dynamic Range
measure of the quietest to the loudest sound it can handle/produce

SNR
Signal to Noise Ratio
-typically want to maximize so that you have all signal, no noise

Headroom
how much you leave so you have margin of error and can deal with unexpected

Frequency Response Graph
graph of the relative amplitude of a range of frequencies

Balanced vs Unbalanced Lines
Balanced - 3 conductors - separated by 2 insulators
TRS Cable - Tip, Ring, Sleeve

Unbalanced - 2 conductors - 1 insulator
TR - Tip and Ring

One of conductors is a ground

Why Bal vs Unbal lines?
Cable are antennas - pick up a ton of noise therefore unbalanced lines pick up noise in the output
Balanced line reverses polarity in one line to cancel out noise

What is the Standard Reference for Pro Gear?
Standard Studio Reference for 0 on your meter is +4dBu
(pro audio strength that will make you VU meter say 0)
{semi pro at -10 dBu}

Transducer
something that converts one kind of energy into another kind of energy

Two classes of microphones -
way they derive the signal - how they turn acoustical energy into electrical
1. Dynamic Microphones
2. Condenser/Capacitor Microphone

Dynamic Microphones
works by the movement of a conductor in a magnetic field

Moving Coil Dynamic
Dynamic Microphone-Diaphragm moves with sound waves that moves coil of wire which is surrounded by a magnet. Therefore voltage is induced in the coil

Ribbon Microphone
Also Dynamic
The diaphragm is the conductor. Can be tiny but it is still a conductor in a magnetic field

Condenser/Capacitor Microphone
Heart has 2 parts: fixed back plate that does not move and a diaphragm that moves with sound wave
Capacitor holds a charge
Think of as valve or pipe
The distance the diaphragm is from the backplate determines how much electricity will pass
Have to send voltage to it (steady voltage in, varying voltage out)

Directional Sensitivity. Why important?
in what direction will it be sensitive to sound and in what direction will it reject sound.
Microphones cannot concentrate on sound therefore we design them with directional sensitivity

3 most common polar patterns
Cardioid - heart shaped - Unidirectional

Bidirectional or figure 8
Equally sensitive front and back (0°, 180°)
Rejects like crazy from sides (90°, 270°)

Omnidirectional - all the directions
Equally sensitive in all directions

How Omnidirectional Polar Pattern is achieved
it is a pressure design
The diaphragm is one side of a sealed enclosure
Microphone doesn't know the direction, it just knows the difference in pressure outside and inside of the enclosure

How Bidirectional Polar Pattern is achieved
the diaphragm is suspended between two anchors and is wide open on both sides
Pressure gradient

How Cardioid Polar Pattern is achieved
pressure gradient plus phase shift
Must have phase shift to work
Sound waves from front hit diaphragm in different wavelengths therefore there is phase shift and it works
Sound waves from back hit diaphragm in the same wavelength therefore there is no phase shift and it does not work

If you cover up these side entry points, the mic will turn into an omni directional mic

Proximity Effect
What it is and what microphones have it?
What it is? - you get more low frequency output out of the microphone the closer it is to the sound source
What microphones? - only pressure gradient design microphones have this
-cardioid and bidirectional
Low frequencies only, high ones are not effective

Loudspeakers
Transducer that takes electrical energy and transforms it into acoustical energy

How loudspeakers work?
Exactly like microphones but reverse

What is a point source?
-theoretical sound source
-omnidirectional radiator
-omnidirectional sound source

Free Space
no barriers, (walls, floor, etc)

A point source is in free space
all points equidistant, acoustical energy is the same

Add 1 barrier (wall - doubling of energy)
3 dB increase (point source hasn't changed)

Add a floor (a wall, a floor, point source in corner)

Another Wall?
6 dB increase, however, point source has still not changed

3dB more again (9dB total) point source doesn't change but we perceive as almost twice as loud

Speakers are tested in
anechoic chambers
anechoic meaning no echos
nothing with point source happen in the chamber, therefore, important to know how room effects sound

Speakers are inefficient. How do we fix?
Put diaphragm in a box, port or vent, and/or have a passive radiator

Passive Radiator
has no electricity- just diaphragm that is powered by the air pressure in the enclosure

Impedance
Speaker has high impedance - takes a lot energy to move it
Air has low impedance - takes little energy to move it
Problem is a mismatch of impedance (like bike in low gear down a hill, speaker doesn't have a lot of air to manipulate)
We want to increase impedance of air to match impedance of speaker

Horn Loading (concert setting)
diaphragm pushes air into a narrower "throat" of a horn
increases impedance of air to match impedance of speaker
make it much louder (efficient)
Folded horn is collapsed inside of an enclosure
If big enough horn, low frequencies could become directional

Frequency vs Driver size
Lower frequencies more effectively produced with larger drivers (diaphragms)
Higher frequencies produced with smaller drivers

Cross over
device that splits the audio into frequency ranges to be assigned to the appropriate driver
-high end often to "tweeters"
-low end to "woofers" (subs = "sub woofers")

Active vs Passive Crossovers
If splitting happens after the power amplifiers, it is a passive crossover
If crossover happens before the amplifiers, it is an active crossover

Why is active more efficient for pro sound and essential for live systems?
can stack up the power on the low frequencies
want the flexibility to distribute amps where you want them
majority of power goes to low ends, we don't struggle to hear high frequencies as much so they don't need as much power

Analog Recording and Types
the storing of a waveform that is analogous to the waveform in the air
Types - vinyl records
Magnetic tape

Magnetic Tape
tape is a long piece of plastic with crushed up magnetic material stuck to it
use transducers to transfer electrical energy into magnetic energy

How Does Tape Recording works?
Wire wrapped around metal, varying voltage sent through it, produces a varying magnetic field, imparts onto tape (record of how field changed over time)

Head produces a varying magnetic field (which is recorded to tape) which is analogous to the change in air pressure

We don't record sound, we record the varying magnetic field which is analogous to the sound

Varying magnetic field is analogous to the varying voltage - which is analogous to the sound (change in air pressure)

3 heads of a professional analog tape machine
Erase
Record
Playback

We want a linear transfer of energy. The problem?
the problem is you don't have a linear transfer from what is applied to the tape to what remains on the tape

Zero Cross Distortion
any magnetic field strength near 0 becomes 0

Bias (or Bias Tone)
a very high frequency (about 150kHz) we add to all analog recordings to get rid of zero cross distortion

Tape Saturation
when the tape can't hold a stronger field because all the magnets are already magnetized, it is saturized
-you can send it a stronger signal but there is not anything to record it
-a kind of distortion that is pleasing for some sounds

Gap Loss Frequency
when the wavelength is short enough at a high frequency so there is no difference between one side of the gap of the playback head and the other
-when no difference, sound drops out, zero output

Why do we want to solve Gap Loss Frequency? What is solution?
will be able to get higher frequencies and record above range of human hearing so we don't hear the distortion
Smaller gap - can't change gap, have to buy them smaller
Move the tap faster- faster the medium, longer the wavelength

Professional Tape Speeds
30 ips or 15 ips

Print Through and how would avoid?
one winding of tape affecting the adjacent winding of the tape (tends to move outward)
-would avoid by not letting them lie on each other
-that is not practical

Cross Talk
very closely related to Print Through
-bleeding one track onto its adjacent
-same section of tape, not windings

Wow and Flutter
Variations in tape speed
Slow variations = wow
Fast variations = flutter
-want as little as possible so you can stay as true to the sound as possible
-want to minimize but it is impossible to completely get rid of