SIASmaart School

1
Dual-Channel FFT Analyzers A Presentation
Prepared for AES-LA Chapter
2
Presenter

• Jamie Anderson
• SIA Product Manager
• Jamie_at_SIASoft.com

SIA Software Company, Inc One Main
Street Whitinsville, MA 01588 508.234.9877 www.sia
soft.com
3
Any idiot can get squiggly line to appear on an
analyzer screen. Our goal is to make ones we
can make decisions on.
Remember Computers do what we tell them to do,
not what we want them to do.
4
To use an analyzer, we must first

• Verify that we are making our measurements
  properly.
• Verify that it is an appropriate measurement for
  our purpose.

5
An analyzer is only a tool YOU make the
decisions

• You decide what to measure.
• You decide which measurements to use.
• You decide what the resulting data means.
• And you decide what to do about it.

6
Example

• The speaker you are measuring has a relatively
  flat response except
• HF is 6 dB lower than LF above 1.6 kHz
• Substantial HF noise

6dB
0dB
1k
2k
4k
8k
500
250
125
-6dB
7
Example
Potential Solutions

• The speaker you are measuring has a relatively
  flat response except
• HF is 6 dB lower than LF above 1.6 kHz
• Substantial HF noise

EQ
Bring up HF on EQ Add extra HF to program
material
6dB
0dB
1k
2k
4k
8k
500
250
125
-6dB
8
Example
Potential Solutions

• The speaker you are measuring has a relatively
  flat response except
• HF is 6 dB lower than LF above 1.6 kHz
• Substantial HF noise

Level
Turn up HF on Amp Turn up HF at X-over output
6dB
0dB
1k
2k
4k
8k
500
250
125
-6dB
9
Example
Potential Solutions

• The speaker you are measuring has a relatively
  flat response except
• HF is 6 dB lower than LF above 1.6 kHz
• Substantial HF noise

System Maintenance
Loss of 6 dB and in crease in HF noise a good
indication of a bad cable on X-over HF out.
6dB
0dB
1k
2k
4k
8k
500
250
125
-6dB
10
Our goal is to fix our systemnot the trace on
the screen.
11
Question How can a X00 piece of software be as
accurate as a XX,000 analyzer?
12
What makes an analyzer?

• Mathematical engine
• Interface
• Hardware
• Engineer

13
Fast Fourier TransformsOur Friend the FFT
14
The Fourier Transform

• Jean Baptiste Joseph Fourier
• All complex waves are composed of a combination
  of simple sine waves of varying amplitudes and
  frequencies

Amp vs Time to
Amp vs Freq
Waveform to
Spectrum
15
Transforms

• A transform converts our data from one domain
  (view) to another.
• Same data
• Is reversible via Inverse Transform
• Yields complex data Magnitude and Phase
  information

Amp vs Time to
Amp vs Freq
Waveform to
Spectrum
16
FFT Resolution

• Reciprocal Bandwidth FR1/TC
• Frequency Resolution 1/Time Constant
• Larger Time Window
• Higher Resolution
• Slower (Longer time window and more data to
  crunch)
• Smaller Time Window
• Lower Resolution
• Faster
• Time Constant Sample Rate x FFT Length
• Decimation Varying SR FFT to get constant
  res.

17
FFT Resolution

• FFTs yield linear data
• Constant bandwidth instead of constant Q
• FFT data must be banded to yield
  fractional-octave data.
• FFT must be windowed
• FFTs assume data is continuous repeating so
  wave form must begin and end at 0.
• Windows are amplitude functions on data

18
FFT Data Windows
19
Math You Need

• T1/
• 1/T

250 Hz T 4 ms
100 Hz T 10 ms
T .1 ms 10 kHz
T .1 ms 10 kHz
20 Hz T 50 ms
T 1 ms 1 kHz
T .5 ms 2 kHz
500 Hz T 2 ms
20
Systems
Input
Output
System
Note These systems can be anything from a
single piece of wire to a multi-channel sound
system with electrical, acoustic and
electro-acoustic elements, as well as wired and
wireless connections. And remember, it only
takes one bad cable to turn a 1,000,000 sound
system into an AM radio!
21
Measurement Types

• Analyzers are our tools for finding problems
• Different measurements are good for finding
  different problems

22
Measurement Types Single Channel vs.
Dual Channel
A()
B()
H()

• Single Channel Absolute
• Dual Channel Relative - In vs Out

Input Signal A () Output Signal B()
Frequency Response H() B()/A()
23
Measurement TypesSingle Channel

• SPL VU
• Wave Form
• Amplitude vs. Time
• Spectrum
• Amplitude vs. Frequency

24
Measurement Types Dual Channel

• Transfer Function Frequency Response
• Phase vs. Frequency
• Magnitude vs Frequency
• Impulse Response
• Magnitude vs Time
• Echo structure

25
Transfer Function
System
Output Signal
Input Signal
Measurement Channel (RTA)
Transfer Function
Reference Channel (RTA)
26
Transfer Function
System
Output Signal
Input Signal
Measurement Channel (RTA)
Transfer Function
Reference Channel (RTA)
27
Transfer Function
System
Output Signal
Input Signal
Measurement Channel (RTA)
Transfer Function
Reference Channel (RTA)
28
What do you get if you transform a transfer
function?

• IFT produces impulse response
• Transfer Function . . . To . . . Impulse Response
• So . . . If Frequency Response can be measured
  source independently - so can Impulse Response

29
Dual-Channel FFT Issues
System
Output Signal
Input Signal

• Window Length vs Resolution
• FR 1/TC Source Independence
• Propagation Time
• Linearity
• Noise
• Averaging
• Coherence

30
How Dual-Channel FFT Analyzers Work
System
Input
Output
Measurement Signal
Reference Signal
Wave
31
How Dual-Channel FFT Analyzers Work
System
Input
Output
FFT

Spectrograph
RTA
FFT
Wave
RTA
32
How Dual-Channel FFT Analyzers Work
System
Input
Output
FFT

FFT
Transfer Function (Frequency Resp.)
Wave
RTA
33
How Dual-Channel FFT Analyzers Work
System
Input
Output
FFT
IFT

FFT
Transfer Function (Frequency Resp.)
Impulse Resp.
Wave
RTA
34
Basic Measurement Set-up
Loudspeaker Room
Source
EQ / Processor
Amplifier
Microphone
Computer w/ Stereo line-level input
Mixer
35
Basic Measurement Set-up
EQ/Processor Control
Loudspeaker Room
Source
EQ / Processor
Amplifier
Control Data
Microphone
Computer w/ Stereo line-level input
Mixer
36
System EngineeringThe Serenity Prayer . . .

• Grant me the SerenityTo accept the things I
  cannot change
• Courage to change the things I can,
• And Wisdom to know the difference.

37
(No Transcript)