Transcript Slide 1

Gain structure, Getting it right
• Gordon Moore CTS
• [email protected]
• 1-800-821-1121
Gain
 The change in level imposed on an electrical
signal
 Generally measured in decibels
 3 dB is a doubling of power
 You need 6 dB to make up for moving twice the
distance from a source – a quadrupling of power
Gain
 10 dB changes the level by a magnitude of ten
(10X or 1/10th )
 20dB changes the level by hundred (100X or
1/100th)
 60 db raises the level a thousand (1,000 or
1/1,000th)
Why is Gain needed?
 Microphones generate VERY low voltage levels – measured in
millivolts.
 This very low signal needs to be amplified to around a volt to be
strong enough for long cable runs.
Why is Gain needed?
 Very low levels are susceptible to interference
from electro-magnetic and radio frequency
sources. Amplification makes these signals
much more resistant to these sources.
 Processing works much better and quieter at
higher voltage levels.
Where is the biggest amplifier?
 The microphone preamp has the greatest effect
on your relative signal levels and S/N ratios. It
is, in general, the most amplification you will
apply in most systems.
 Get the preamplifier right and everything else
falls into line
Gain is KING
 Proper gain staging makes or breaks performance at all levels.
 It can:
 Affect signal to noise ratio in the entire system
 Affect the proper operation of signal processors
 DSP
 Conferencing systems
 Audio processors
 Affect wireless microphone performance
What is gain staging
 The optimization of the signal level so all components in
the signal chain are operating within their optimal
performance range without detriment to signal integrity.
 Or
 Setting the levels right…………………..
Signal to noise ratio
Poor S/N
Ratio
Line
Level
Mic
Level
Noise Floor
Mic Preamp Mixer
Signal processors
Amps
Signal to noise ratio
Line
Level
Good
S/N
Ratio
Mic
Level
Noise Floor
Mic Preamp Mixer
Signal processors
Amps
Signal to noise ratio
Line
Level
Mic
Level
Noise Floor
Mic Preamp Mixer
Signal processors
Amps
Two methods to be discussed
 Unity gain staging
 System optimization
Common attributes
 Both, when properly engaged, will provide improved signal to noise
ratio.
 Both require some minor specialized equipment
 Signal generator
 Tone and/or pink noise
 Signal analyzer
 Simple as Multimeter or even cheap piezo tweeter
 Misc cables with all sorts of connector combinations – great soldering
practice
Good news
 – Neither method requires really expensive equipment. You DO
NOT need
 Oscilloscopes, Audio analyzers or
other stuff requiring $$$$$$$$$$$$
Why two methods?
 The correct answer in all questions audio is
IT DEPENDS
Unity Gain Method
 Involves setting all line level outputs to one common level
 Most often 0dBu (.775Volts) or +4dBu (1.94V)
 Use – permanent installed sound systems such as courtrooms,
conference centers, boardrooms, paging etc – primarily voice
reinforcement systems
Unity Gain Method
 Advantage – maintenance is simple –
 pull out device A,
 replace with device B,
 set output level to 0dBu,
 voice test,
 Go home, drink cold beverage
 Unity gain – 0dBu in to 0dBu out.
Unity Gain Method
 Why 0dBu or +4dBu?
 Most systems need “headroom” – generally 20dB will
handle anything
 Most professional gear can handle up to +20dBu
(77.5V) signals (professional theatre or touring gear
up to +28dBu)
 Capping “normal” level at 0 allows up to 20dB of extra
loud signal before distortion or clipping
Clipping
 Driving an amplifier circuit to the point where signal
levels cannot go any higher.
 OVERDRIVEN signals create distortion and harmonics
 Can overheat components and destroy speakers.
 Guitar amplifiers/speakers that must handle distorted
guitars require much heftier components than most
home systems can handle.
 Example
Unity Gain Method
 Requires signal generator – tone and pink noise
 Meter to measure output

http://www.nti-audio.com/Home/Products/DiscontinuedProducts/tabid/97/Default.aspx
Another option
 http://www.world-voices.com/software/nchtone.html
 Or Google “Tone generator free”
Unity Gain Method
 Insert Known signal into input at TOP of chain
 Top = first audio insertion point – microphone input on
mixer
 Known signal can be any level – but you must know
what that level is.
 Set input trim to set 0dBu level
 (if input signal is -20dBu, set input trim at +20 to get 0)
Unity Gain Method
 Set output level to measure 0dBu (or .775V with multimeter set
on AC)
 Go to next device – set input and output gain so level at output is
also 0dBu
 Set amplifiers last – if all else is set right and amps are properly
chosen, typical amp level will be “low”.
Unity Gain Method
 Until you get to amplifiers, there is NO need to
have amplifiers ON.
 When setting amplifier, use typical test signal
and set levels in room using SPL meter
 (or even by ear, not documentable however)
 Equalize then reset equalizer output to 0dBu.
Unity Gain Method
 Basic system gain is correctly set.
 Any changes to any signal processors, (compressors, limiters, etc)
will require a reset of that piece of gear back to unity gain.
Demonstration -
System optimization
 Setting every component so that clipping occurs at exactly the same
time in all devices. If “clip” indicator shows nothing on main console,
you know everything else in your system is not going to “clip” either.
All components can be driven to their maximum without danger.
System Optimization
 WHEN to use? Live sound systems and theatrical installations
where the sound operators may need to drive the system to “Max”.
This optimizes the signal to noise ratio and permits full signal range.
Requires complete re-calibration if any components are replaced.
System optimization
Two ways to do this – one requires expensive gear, the other
doesn’t
Lets go cheap first
You need
CHEAP STUFF
1 – very cheap piezo tweeter – generally available
for less than $10.
2 – one signal generator that can generate an A on
the scale 440Hz (or something near that level)
If you have a laptop, you can get free on-line
If not, music stores have generators for less than $50.
Piezo Speaker
System optimization
 Step one – connect speaker to output of first device
(generally the mixer)
 Step two – set output level at 0, faders at 0, subs at 0.
 Step three – insert tone into input of mixer. Increase
gain at input until speaker makes squealing sound (you
are clipping) – back input gain down until silent.
System optimization
 At this point, leave all controls at mixer alone – do not change
 Repeat test at each input and output of all line level devices
downstream.
 When you are finished, all devices will clip at the same time.
Running up gain until
clipping
System optimization
 Expensive method – replace free laptop
software with fancy signal generator and
speaker with O-Scope.
 Results will be exactly the same.
 If you lose the speaker or it gets stolen – oh well.
 If you lose the O-scope ………….
Demonstration
Setting mics
 Based on application and varies according to mic.
 Set input trim to give 0dBu nominal level
 Can shouting distort the audio (clipping!)?
 If Yes, turn it down a bit until answer is no.
Setting Multi-Media
 Use known reference – as in KNOWN to YOU
 Ideally a measured sample – so you can document the setup
and result
 Does not have to be boring test signals
Setting Multi-Media
 I use Diana Krall, “Live in Paris”
 Reasons –
 Wide dynamic range (soft to loud)
 Wide frequency range (bass to cymbals)
 Available in DVD and CD format
 I also use boring test CD as well. ZZZZZZZZZZZZZZZZZZ
 Pink noise drives out unwanted visitors.
Setting wireless mics
 Transmitter gain affects FM modulation
 The modulation of the signal affect S/N ratio
Poor modulation makes apparent range lower
 Getting transmitter gain “hot” improves FM performance
Setting Wireless Microphones
 Again – avoid distortion but don’t set too low
 Depends on situation – does scene require
screaming? Or simply talking?
 Then set output of receiver according to input in
the mixer you are using.
 Line level preferred – better resistance in RF and EMI
So, why is this important again?
 If gain is consistently set at optimal levels you get three benefits
Benefit 1 –
 – relative levels between sources will be even
 Far side v near side in conferencing
 Dial tones v incoming phone level
 Multi-media v voices
Benefit 2
 Processing functions will work at their best
 Compressors, limiters, automixers, equalizers, etc
Benefit 3
 – Signal to noise ratio will be at best possible result
Summary
 Maximum modulation with good headroom will give best signal to
noise ratio.
 Set ALL stages (don’t forget any inputs)
Questions?