Digital Audio and Class D Amplification
Download
Report
Transcript Digital Audio and Class D Amplification
By: Logan Sealover
Analog-to-Digital Conversion (ADC)
Digital-to-Analog Conversion (DAC)
History of Class D Amplifiers
Audio Amplifier Classes
What Makes it Class D?
Class D Distortion Disaster
Frequent Future
Today’s world—smartphones & laptops
Class D amplifiers currently at the top of the line
◦ Rated at 90% efficiency—less heat
The “D” in Class D amplification
Analog audio signals
Can be any continuous
value
Signals can only
understand binary
Unwanted background
noise
Can compress analog
into smaller files
12” LP Record faults
CD “skips”
Analog signals
◦ Must be converted to binary for use in digital equipment
Phone line
Audio CD
◦ Figure 1—analog audio wave signal
Y-axis—Voltage (V)
X-axis—Time (t)
ADC Sampling
Figure 1
Sampling Rate
Figure 2—sampled analog signal
◦ A sample rate of 1 Hz
uses 1 sample point/sec
◦ 22,050 Hz samples
22,050 points/second
◦ 44,100 Hz samples
44,100 points/second
◦ Space Requirements
34 Hz/second
Nyquist Theorem
◦ Sampling Rate > 2 * Highest recorded frequency (Hz)
◦ Determines the best sampling rate during ADC for the
best storage and sound quality
Human Hearing Range (20 Hz – 20,000 Hz)
◦ Typical music sampling rate (44,100 Hz)
◦ Phone system sampling rate (8,000 Hz or 8 kHz)
Sample point value and bit size
◦ n-bits
2n = N available states
n = 8 bits
n = 16 bits
◦ 0 – N defines the frequency of the sample point
◦ Greater sample point size means better sound quality,
but also means more storage space is needed
Signal-to-Noise Ratio (SNR)
◦ SNR = 6.02 * n (# of sample point bits) + 1.76 dB
◦ Calculates the desired noise level of your audio
application’s tolerable noise level
◦ Higher SNR provides better quality
Sample Point Bit Resolution Sizes
◦ 8 bit – Phone Systems
◦ 16 bit – Audio CDs
◦ 20 and 24 bit – High-end DVD audio
Ex:) Coheed & Cambria – Welcome Home (8 bit)
http://www.youtube.com/watch?v=ggahA5kQjHI
With the sampling rate and sample point bit size,
we can determine the necessary storage space
Phone System Quality:
◦ Sampling Rate – 8,000 Hz (8 kHz)
◦ Sample Point Size – 8 bits (1 byte)
◦ Transmission Rate: 8,000 Hz * 8 bits = 64,000 bps
= 8,000 bytes/second OR 480,000 bytes/minute
Audio CD Quality:
◦
◦
◦
◦
Sampling Rate – 44,100 Hz
Sample Point Size – 16 bits
Two Independent Channels – Left & Right
44,100 Hz * 16 bits * 2 channels = 176,400 bytes/second
OR 10,584,000 bytes/minute (~10 MB/minute)
◦ 720 MB of available space = 72 minutes
CD-ROM Quality:
◦ Storage space is slightly less due to error-correction code
◦ 650 MB of available space
DAC converts binary number
patterns into voltages and
currents for your speakers
The DAC only connects
the points that were captured
by the previous ADC
◦ Audio signal not always the exact
same as it was recorded
◦ Skipped values without curves
Figure 3
Much like a reverse process of ADC
Suffers sound quality loss
Digital-to-Analog Converters are unavoidably
expensive
SHARP SM-SX1
Pulse-Width Modulation
For decades now, Class D amplifiers have been
used in devices where high efficiency is important
◦ Medical field—Hearing aids
◦ Large controllers for bulky motors and electromagnets
Recently released to public
◦ Tripath Technology, Texas Instruments, Cirrus Logic, etc.
MP3/CD players
Laptops
Cellphones/PDAs
Home audio (TVs and stereos)
Class D amplifiers ALWAYS have their transistors
operating either fully on, or fully off
Able to accept a stream of bits from a CD/MP3
player and convert it to an analog signal
Older models are entirely analog
◦ Amplify digital signals only after conversion to analog
◦ Figure 5
Figure 5
Different topologies and classes depending on
how much current is allowed to pass while passive
Common Designs:
◦
◦
◦
◦
Class A
Class B
Class AB
Class D
Other Designs:
◦ Class C
◦ Class E & F
◦ Class G & H
Class A
◦ No crossover distortion
◦ Wastes 50% of power
◦ Excess heat
Class B
◦ Crossover region near 0 V
◦ Push-Pull one at a time
◦ Two transistors
Class B cont.
◦ Crossover distortion
◦ Fairly efficient—78% of power used
◦ Remaining power dissipated as heat
Class AB
◦ Push-Pull simultaneously
◦ Two transistors
◦ Smoother transfer rate
Less distortion
◦ Lower efficiency than B
Class D uses Push-Pull between two transistors
◦
◦
◦
◦
Switches between two voltage values (e.g., ±40 V)
Can connect the output to both transistors simultaneously
Neither transistor wastes any power
Binary wave signals must be converted (DAC)
Other Amplifier Designs:
◦ Class C amps are used only for radio frequencies
◦ Class E & F are used for higher radio frequencies
◦ Class G & H amps use more complex variations
of other amp classes for specialized applications
Class D is prone to distortion
◦ Imperfect power supply regulation
◦ Timing errors
Power Supply Modulations caused by variations
in the amount of current drawn by the amplifier
◦ Extra noise, or hum, from power supply fluctuations
Timing Errors due to changes in how long the
transistors take to switch from on to off
Frequency Response is the accuracy and
equality of the sounds being produced during DAC
◦ Helps keep different frequencies at equal volume levels
These problems can be fixed using an analog
feedback system to compensate for output-stage
distortion
◦ Some of these systems handle frequency-response
problems too
Research will increasingly focus on Digital Signal
Processing to correct inevitable analog errors
◦ Controllers that can sense voltage and modify their
signals accordingly
Circuits that perform digital modulation by
measuring analog error data and modifying the
switch control signal as a result
Digital sampling rate and its quality/storage
ADC/DAC not perfect
Amplifier classes
Class D distortion
Future prospects
“Amplifier.” Wikipedia. Wikimedia Foundation, 13
Feb. 2013. Web. 17 Feb. 2013.
Putzeys, B. "Digital Audio's Final Frontier." IEEE
Spectrum 40.3 (2003): 34-41. Print.
Torres, Gabriel. "How Analog-to-Digital Converter
(ADC) Work." Hardware Secrets. 21 Apr. 2006.
Web. 18 Feb. 2013.
END
(Applause)