TI Class D Amplifier

Application Report
SLOA068 October 2001
Guidelines for Measuring Audio Power Amplifier

Performance
Audio Power Amplifiers
ABSTRACT
This application note provides guidelines for measuring the data sheet parameters of
Texas Instruments audio power amplifiers (APAs) using prefabricated evaluation modules
(EVMs). The primary equipment used for the measurements consists of the System
Two audio measurement system by Audio Precision, a digital multimeter (DMM), and
a dc power supply.
Contents
1 Introduction .....................................................................................................................................2
2 Basic Measurement System...........................................................................................................3
3 Interfacing to the APA.....................................................................................................................5
3.1 Differential Input and BTL Output (TPA731 and TPA2000D1) ..................................................5
3.2 SE Input and SE Output (TPA0211 and TPA711).....................................................................6
3.3 Other Configurations .................................................................................................................7
3.4 Class-D RC Low-Pass Filter......................................................................................................7
4 Total Harmonic Distortion Plus Noise (THD+N) ...........................................................................9
4.1 THD+N vs Output Power.........................................................................................................10
4.2 THD+N vs Frequency..............................................................................................................11
4.3 Maximum Output Power Bandwidth ........................................................................................11
4.4 Maximum Input Voltage...........................................................................................................11
5 Noise ..............................................................................................................................................12
5.1 Integrated Noise vs Frequency ...............................................................................................12
5.2 Signal-to-Noise Ratio ..............................................................................................................13
6 Gain and Phase .............................................................................................................................13
7 Crosstalk ........................................................................................................................................15
8 Supply Rejection ...........................................................................................................................17
9 Power Measurements and Related Calculations........................................................................21
9.1 Efficiency Measurements ........................................................................................................21
9.2 Power Dissipated vs Power to the Load..................................................................................24
9.3 Crest Factor and Output Power...............................................................................................25
10 Measurement Pitfalls ....................................................................................................................26
10.1 Effects of Improper Interfacing and Grounding .......................................................................26
10.2 THD+N Measurements............................................................................................................27
10.3 Noise Measurements ..............................................................................................................27
10.4 Gain and Phase Measurements ..............................................................................................28
10.5 Crosstalk Measurements.........................................................................................................28
10.6 Supply Rejection Measurements.............................................................................................28
10.7 Efficiency Measurements ........................................................................................................28
11 References .....................................................................................................................................28
Audio Precision and System Two are trademarks of Audio Precision, Inc.
Other trademarks are the property of their respective owners.
1
SLOA068
Figures
Figure 1. Audio Measurement Systems: (a) Class-AB APAs and (b) Filter-Free Class-D APAs..... 4
Figure 2. Differential InputBTL Output Measurement Circuit ......................................................... 5
Figure 3. SE InputSE Output Measurement Circuit ......................................................................... 7
Figure 4. Measurement Low-Pass Filter Derivation CircuitClass-D APAs .................................... 8
Figure 5. THD+N Measurement Circuit Using the AP-II Measurement System:
Differential-BTL.................................................................................................................... 10
Figure 6. THD+N vs POUT for the TPA2001D1 and the TPA731 ......................................................... 10
Figure 7. THD+N vs Frequency for the TPA2001D1 and the TPA731 .............................................. 11
Figure 8. Noise Measurement Circuit ................................................................................................. 12
Figure 9. Measured Results of Noise Circuit ..................................................................................... 13
Figure 10. Gain and Phase Measurement Circuit ................................................................................ 14
Figure 11. TPA731 Gain and Phase Measurements ............................................................................ 14
Figure 12. TPA2001D1 Gain and Phase Measurements...................................................................... 15
Figure 13. Crosstalk Measurement Circuit........................................................................................... 16
Figure 14. Crosstalk Measurements ..................................................................................................... 17
Figure 15. PSRR and kSVR Measurement Circuit.................................................................................. 18
Figure 16. kSVR Filter Circuit................................................................................................................... 19
Figure 17. kSVR of the TPA2001D1 and TPA731.................................................................................... 20
Figure 18. Impact of CBYPASS on kSVR for the TPA711 Class-AB APA ................................................. 20
Figure 19. Efficiency Measurement Circuit for Class-AB and Class-D BTL APAs........................... 22
Figure 20. Efficiency Graphs of the TPA731 and TPA2001D1............................................................ 24
Figure 21. Graph of Power Dissipated vs Output Power .................................................................... 24
Figure 22. Supply and Output Power vs CF for the TPA731 and TPA2001D1 .................................. 26
Figure 23. Effect of Generator Interface on APA Measurements, THD+N vs Power Shown ........... 27
Tables
Table 1. Recommended Minimum Wire Size for Power Cables............................................................. 6
Table 2. Typical RC Measurement Filter Values ..................................................................................... 9
Table 3. Efficiency Data for the TPA731 and TPA2001D1 .................................................................... 23
Table 4. Power vs Crest Factor............................................................................................................... 25
1 Introduction
The primary goal of audio measurements is to determine the performance of a device in the
audible spectrum, 20 Hz to 20 kHz. Although most people do not hear frequencies below 50 Hz
or above 17 kHz, the broad spectrum is an industry standard that allows a more accurate
comparison of devices. The performance can be quickly analyzed, and only a few basic pieces
of equipment are required.
A method for measuring standard data sheet information for audio power amplifiers (APAs) is
presented for several key parameters. These are:
THD+N versus output power Crosstalk versus frequency

THD+N versus frequency Power supply rejection ratio
Gain and phase versus frequency Supply ripple voltage rejection ratio
Integrated noise Efficiency
Signal-to-noise ratio Power dissipated in the device
2 Guidelines for Measuring Audio Power Amplifier Performance

SLOA068
The measurements in this application note were made using TI Plug-N-Play APA evaluation
modules (EVMs). The TPA2001D1 and TPA731 mono devices were used for most
measurements. The TPA2001D2 and TPA0212 devices were used for the crosstalk
measurements, which require a stereo device.
Note that the measurements are dependent upon the layout of the printed-circuit board (PCB),
particularly with class-D APAs. The graphs in the data sheet reflect typical specifications and
were measured on test boards specifically designed to allow accuracy and ease of
measurement. The measurements in this application note, however, were taken using circuits on
EVMs that reflect real-world layout constraints. The measurements of a particular audio circuit
may vary from the typical specifications. A large variance is usually indicative of a PCB layout or
measurement system issue.
2 Basic Measurement System

This application note focuses on methods that use the basic equipment listed below:
Audio analyzer or spectrum analyzer Digital multimeter (DMM)

Oscilloscope Twisted pair wires
Signal generator Power resistor(s)
Linear regulated power supply Filter components
EVM or other complete audio circuit
Figure 1 shows the block diagrams of basic measurement systems for class-AB and class-D
amplifiers. A sine wave is normally used as the input signal since it consists of the fundamental
frequency only (no other harmonics are present). An analyzer is then connected to the APA
output to measure the voltage output. The analyzer must be capable of measuring the entire
audio bandwidth. A regulated dc power supply is used to reduce the noise and distortion injected
into the APA through the power pins. A System Two audio measurement system (AP-II)
(Reference 1) by Audio Precision includes the signal generator and analyzer in one package.
The generator output and amplifier input must be ac-coupled. However, the EVMs already have
the ac-coupling capacitors, (CIN), so no additional coupling is required. The generator output
impedance should be low to avoid attenuating the test signal, and is important since the input
resistance of APAs is not very high (about 10 k). Conversely the analyzer-input impedance
should be high. The output impedance, ROUT, of the APA is normally in the hundreds of milli-
ohms and can be ignored for all but the power-related calculations.
Figure 1(a) shows a class-AB amplifier system, which is relatively simple because these
amplifiers are lineartheir output signal is a linear representation of the input signal. They take
analog signal input and produce analog signal output. These amplifier circuits can be directly
connected to the AP-II or other analyzer input.
This is not true of the class-D amplifier system shown in Figure 1(b), which requires low pass
filters in most cases in order to measure the audio output waveforms. This is because it takes an
analog input signal and converts it into a pulse-width modulated (PWM) output signal that is not
accurately processed by some analyzers.
Guidelines for Measuring Audio Power Amplifier Performance 3

SLOA068
Power Supply
Signal APA RL Analyzer

Generator 20 Hz - 20 kHz
(a) Basic Class-AB Audio Measurement System
Power Supply
Low-Pass RC
Filter
Signal Class-D APA Low-Pass RL Analyzer
Generator LC Filter 20 Hz - 20 kHz
Low-Pass RC
Filter
Used With Traditional Class-D APAs Only

Used With Filter-Free Class-D APAs Only
(b) Filter-Free and Traditional Class-D Audio Measurement System
Figure 1. Audio Measurement Systems: (a) Class-AB APAs and (b) Filter-Free Class-D APAs
Two types of class-D amplifiers exist: traditional class-D that requires a low-pass LC filter to
produce an analog output, and TIs new filter-free class-D which does not require a low-pass
output filter for normal operation because the speaker provides the inductance necessary to
achieve high efficiency.
Two families of class-D APAs (TPA032D0x, TPA005Dxx) use the traditional modulation scheme
that requires the LC filter for proper operation. The data sheets, EVM manuals, and application
notes (References 2 and 3) provide more information about this filter.
The filter-free class-D APA families (TPA2000Dx and TPA2001Dx) use a modulation scheme
that does not require an output filter for operation, but they do sometimes require an RC low-
pass filter when making measurements. This is because some analyzer inputs cannot accurately
process the rapidly changing square-wave output and therefore record an extremely high level of
distortion. The RC low-pass measurement filter is used to remove the modulated waveforms so
the analyzer can measure the output sine wave.

SLOA068
3 Interfacing to the APA
This section describes the important points to be considered when connecting the test
equipment to the APA. The first two subsections describe the connections to differential and
single-ended (SE) APA inputs and outputs. The last subsection discusses the RC low-pass filter
design that is sometimes required for filter-free class-D measurements.
3.1 Differential Input and BTL Output (TPA731 and TPA2000D1)

All of the class-D APAs and many class-AB APAs have differential inputs and bridge-tied load
(BTL) outputs. Differential inputs have two input pins per channel and amplify the difference in
voltage between the pins. Differential inputs reduce the common-mode noise and distortion of
the input circuit. BTL is a term commonly used in audio to describe differential outputs. BTL
outputs have two output pins providing voltages that are 180 degrees out of phase. The load is
connected between these pins. This has the added benefits of quadrupling the output power to
the load and eliminating a dc blocking capacitor.
A block diagram of the measurement circuit is shown in Figure 2. The differential input is a
balanced input, meaning the positive (+) and negative (-) pins will have the same impedance to
ground. Similarly, the BTL output equates to a balanced output.
Evaluation Module
Audio Power
Generator Analyzer
Amplifier
CIN Low-Pass
RC Filter
RGEN RIN ROUT RANA CANA
VGEN RL
CIN
Low-Pass
RGEN RIN ROUT RC Filter RANA CANA
Twisted-Pair Wire Twisted-Pair Wire

The RC low-pass filter is required only for measuring the filter-free class-D audio power amplifiers.
Figure 2. Differential InputBTL Output Measurement Circuit
The generator should have balanced outputs and the signal should be balanced for best results.
An unbalanced output can be used, but it may create a ground loop that will affect the
measurement accuracy. The analyzer must also have balanced inputs for the system to be fully
balanced, thereby cancelling out any common mode noise in the circuit and providing the most
accurate measurement.
The following general rules should be followed when connecting to APAs with differential inputs
and BTL outputs:
Use a balanced source to supply the input signal.
Use an analyzer with balanced inputs.
Use twisted-pair wire for all connections.

SLOA068
Use shielding when the system environment is noisy.
Ensure the cables from the power supply to the APA, and from the APA to the load, can
handle the large currents (see Table 1 below).
Table 1 shows the recommended wire size for the power supply and load cables of the APA
system. The real concern is the dc or ac power loss that occurs as the current flows through the
cable. These recommendations are based on 12-inch long wire with a 20-kHz sine-wave signal
at 25C.
Table 1. Recommended Minimum Wire Size for Power Cables
PO UT(W) RL () AWG DC Power AC Power

Size Loss (mW) Loss (mW)
10 4 18 16 18
22 40 42
2 4 18 3.2 3.7
22 8.0 8.5
1 8 22 2.0 2.1
28 8.0 8.1
< 0.75 8 22 1.5 1.6
28 6.1 6.2
3.2 SE Input and SE Output (TPA0211 and TPA711)

The SE input and output configuration is used with class-AB amplifiers only. A block diagram of
a fully SE measurement circuit is shown in Figure 3. Fully SE APAs are, in general, headphone
or headset amplifiers, though the TPA0211 and TPA711 are APAs with SE capability. SE inputs
normally have one input pin per channel. In some cases two pins are present; one is the signal
and the other is ground. SE outputs have one pin driving a load through an output ac coupling
capacitor and the other end of the load is tied to ground. SE inputs and outputs are considered
to be unbalanced, meaning one end is tied to ground and the other to an amplifier input/output.
The generator should have unbalanced outputs, and the signal should be referenced to the
generator ground for best results. Unbalanced or balanced outputs can be used when floating,
but they may create a ground loop that will effect the measurement accuracy. The analyzer
should have balanced inputs to cancel out any common-mode noise in the measurement.

SLOA068
Evaluation Module
Audio Power
Generator Analyzer
Amplifier
CIN
CL
RGEN RIN
VGEN
ROUT RANA CANA
RL
RANA CANA
Figure 3. SE InputSE Output Measurement Circuit
The following general rules should be followed when connecting to APAs with SE inputs and
outputs:
Use an unbalanced source to supply the input signal.
Use twisted pair wire for all connections.
Ensure the cables from the power supply to the APA, and from the APA to the load, can
handle the large currents (see Table 1, Section 3.1)
3.3 Other Configurations

Some APAs are designed to operate in some combination of the two previously discussed
configurations. For example, the TPA0312 is configured with differential inputs and SE outputs
while the TPA711 is configured with SE inputs and BTL outputs. The TPA0212 can be operated
with any combination of inputs and outputs. The relevant portions of Sections 3.1 and 3.2 are
then used to configure the measurement system properly.
3.4 Class-D RC Low-Pass Filter

An RC filter is used to reduce the square-wave output when the analyzer inputs cannot process
the pulse-width modulated class-D output waveform. This filter has little effect on the
measurement accuracy because the cutoff frequency is set above the audio band. The high
frequency of the square wave has negligible impact on measurement accuracy because it is well
above the audible frequency range and the speaker cone cannot respond at such a fast rate.
The RC filter is not required when an LC low-pass filter is used, such as with the class-D APAs
that employ the traditional modulation scheme (TPA032D0x, TPA005Dxx).

SLOA068
The component values of the RC filter are selected using the equivalent output circuit as shown
in Figure 4. RL is the load impedance that the APA is driving for the test. The analyzer input
impedance specifications should be available and substituted for RANA and CANA. The filter
components, RFILT and CFILT, can then be derived for the system. The filter should be grounded
to the APA near the output ground pins or at the power supply ground pin to minimize ground
loops.
Load RC Low-Pass Filters AP Analyzer Input

RFILT
CFILT CANA RANA
RL VL= VIN
VOUT
RFILT
CFILT CANA RANA

To APA
GND
Figure 4. Measurement Low-Pass Filter Derivation CircuitClass-D APAs
The transfer function for this circuit is shown in Equation (1) where O = REQCEQ, REQ =
RFILTRANA and CEQ = (CFILT + CANA). The filter frequency should be set above fMAX, the highest
frequency of the measurement bandwidth, to avoid attenuating the audio signal. Equation (2)
provides this cutoff frequency, fC. The value of RFILT must be chosen large enough to minimize
current that is shunted from the load, yet small enough to minimize the attenuation of the
analyzer-input voltage through the voltage divider formed by RFILT and RANA. A rule of thumb is
that RFILT should be small (~100 ) for most measurements. This reduces the measurement
error to less than 1% for RANA 10 k.
R ANA

VOUT R ANA + R FILT
= (1)
VIN
1 + j

O
f C = 2 fMAX (2)
An exception occurs with the efficiency measurements, where RFILT must be increased by a
factor of ten to reduce the current shunted through the filter. CFILT must be decreased by a factor
of ten to maintain the same cutoff frequency. See Table 2 for the recommended filter component
values.
Once fC is determined and RFILT is selected, the filter capacitance is calculated using
Equation (3). When the calculated value is not available, it is better to choose a smaller
capacitance value to keep fC above the minimum desired value calculated in Equation (2).

SLOA068
1
C FILT = (3)
2 f C R FILT
Table 2 shows recommended values of RFILT and CFILT based on common component values.
The value of fC was originally calculated to be 28 kHz for an fMAX of 20 kHz. CFILT, however, was
calculated to be 57 000 pF, but the nearest values of 56 000 pF and 51 000 pF were not
available. A 47 000 pF capacitor was used instead, and fC is 34 kHz, which is above the desired
value of 28 kHz.
Table 2. Typical RC Measurement Filter Values

Measurement RFILT CFILT
Efficiency 1 000 5 600 pF
All other measurements 100 56 000 pF
4 Total Harmonic Distortion Plus Noise (THD+N)

The THD+N measurement combines the effects of noise, distortion, and other undesired signals
into one measurement and relates it (usually as a percentage) to the fundamental frequency.
Ideally, only the fundamental frequency of the sine-wave input is present at the output of the
APA, which in practice is never the case. Nonlinearities in the APA, internal and external noise
sources, and layout or grounding issues are some of the contributors that distort the original
input signal.
THD+N requires measuring the value of everything that remains, which includes harmonics and
noise, after the fundamental frequency has been filtered. This value is then divided by the
fundamental frequency and expressed as a percentage. The bandwidth is often limited to record
only the portion of the noise in the audible spectrum. The signal generator, audio analyzer, and
filters should have a noise floor and distortion that is at least 10 dB lower than the APA distortion
in order to achieve an accurate measurement (Reference 4).
Figure 5 shows an Audio Precision II (AP-II) system setup for measuring the THD+N of
differential-BTL APAs. The bandwidth is usually limited with filters in the analyzer to reduce the
out-of-band noise; however, this also reduces relevant harmonics of the higher frequency
signals. A filter cutoff frequency of 30 kHz is used for class-AB and class-D APAs to allow
measurement of the third harmonic for a 10 kHz signal. The narrow bandwidth attenuates the
distortion at higher frequencies, but these harmonics are beyond the audible threshold of the
human ear and are not a factor.
Three measurements that express THD+N in some manner in the data sheets are THD+N
versus output power, THD+N versus frequency, and the maximum output power bandwidth,
covered respectively in the following Sections 4.1 through 4.3. Section 4.4 provides a means to
calculate and measure the maximum input voltage for an APA. These measurements vary with
CBYPASS for devices that have a BYPASS pin, with THD+N increasing as CBYPASS decreases.

SLOA068
AP Generator Out Audio Power Amplifier AP Analyzer In
CIN RFILT
+ IN+ OUT+ +
Diff Inputs
RL CFILT
Channel A CIN BTL RFILT Channel A
-
IN- Outputs OUT-
-
CFILT Inputs Balanced

THD+N vs POUT: VS GND AC-Coupled
Outputs Balanced
Zin = 100 k/185 pF
Zout = 40
Set Load Reference = RL
Internal Filter = 30 kHz
Sweep 10 mW - POUT(max)
Low-Pass RC Reading Meter = THD+N
Fixed Frequency
Filter for Ratio
V+ GND
THD+N vs Frequency: Class-D
Outputs Balanced Measurements
Zout = 40
Set Load Reference = RL Regulated
Sweep 20 Hz - 20 kHz Power Supply
Fixed Amplitude
Figure 5. THD+N Measurement Circuit Using the AP-II Measurement System:

Differential-BTL
4.1 THD+N vs Output Power

A graph of THD+N versus output power is shown in Figure 6. The signal generator sweeps the
input voltage from low to high amplitude at a fixed frequency. The output power is calculated for
a given load impedance that is entered into the audio analyzer software. At each voltage step
the fundamental frequency is measured first, then filtered out and the amplitude of all the
remaining harmonics is measured. This value is then divided by the amplitude of the
fundamental frequency and graphed as a percentage of the fundamental.
The higher distortion at low values of POUT is due to the decrease in signal-to-noise ratio as the
harmonics decrease in amplitude below the noise floor (Reference 4). The sudden increase at
the upper level of POUT is due to clipping of the output signal.
2.0
VS = 3.3 V AV = 12 dB Class-AB
RL = 8 AV = 6 dB Class-D
CB = 1 F f = 1 kHz
BTL
THD+N (%)
0.2
Class-AB
Class-D
0.02
0.01 0.1 1.0
POUT (W)
Figure 6. THD+N vs POUT for the TPA2001D1 and the TPA731

SLOA068
4.2 THD+N vs Frequency
A graph of THD+N versus frequency is shown in Figure 7. The signal generator sweeps the
frequency from 20 kHz to 20 Hz at a fixed voltage. The harmonics and noise of the APA output
are measured at specified frequency steps. Each step is divided by the amplitude of the
fundamental frequency and graphed as a percentage of the fundamental. This graph provides a
check when compared to the THD+N versus power since they should match at one specific
frequency and power.
The increased THD+N at low frequencies is primarily due to the 1/f noise. The high frequency
THD+N increase is due to device nonlinearities, primarily crossover distortion, and is expected
because the APA open loop gain decreases with frequency. The audio quality is unaffected
because the harmonics are above the audible threshold of the human ear (Reference 5). The
rolloff at high frequencies is due to the band-limiting filter in the analyzer, which attenuates the
upper harmonics above 30 kHz. Setting the filter frequency higher reduces the accuracy of the
measurement with class-D APAs, and will have little or no impact on class-AB APAs. The class-
AB graph continues in a relatively straight line if there is no filter present. The class-D rolls off
more than class-AB because of the RC measurement filter, which adds another pole at 30 kHz.
2.0
RL = 8 6 dB Class-D
CB = 1 F POUT (class-AB) = 250 mW
BTL POUT (class-D) = 300 mW
THD+N (%)
0.2
Class-AB
Class-D
0.02
20 200 2k 20k
Frequency (Hz)
Figure 7. THD+N vs Frequency for the TPA2001D1 and the TPA731
4.3 Maximum Output Power Bandwidth

The maximum output power bandwidth is a THD+N versus frequency measurement. The APA is
driven at the maximum output power into the load and the frequency is swept from 20 Hz to 20
kHz. The maximum power bandwidth is then specified as the frequency range over which the
THD+N remains below a specified percentage, which is normally one percent.
4.4 Maximum Input Voltage

The maximum input voltage required for producing maximum output power can be found by
increasing the input until the output clips, then reducing it until it is just below clipping. Another
method is to calculate the maximum peak-to-peak input voltage using the maximum-rated RMS
output power from the data sheet or back-calculate it from the THD+N versus power
measurement at the maximum desired value of distortion. Equation (4) provides the maximum
peak-to-peak input voltage, where POUT(max) is the maximum rated RMS output power, RL is the
load resistance, and AV is the voltage gain of the APA, measured in V/V.

SLOA068
2 2 POUT(max) R L
VIN(P P ) = (4)
AV
5 Noise
Two types of measurements fall under the noise category, integrated noise over the audio band
and signal-to-noise ratio (SNR) of the output signal.
5.1 Integrated Noise vs Frequency

The noise measurement circuit is shown in Figure 8 for an APA with differential inputs and BTL
outputs. A graph depicting the output noise voltage of the TPA2001D1 and the TPA731 is shown
in Figure 9. All of the inputs of the APA should be ac-coupled to ground through the input
resistor, whether internal or external, to reduce noise pickup and accurately simulate the system.
A graph of THD+N versus POUT is shown in Figure 6. The AP generator outputs are not used in
this measurement and should be turned off.
The analyzer should be set to measure amplitude and should be limited to measure the noise in
the audio spectrum only. The bandwidth is limited to the range of 22 Hz 22 kHz with filters in
the analyzer. The data field of the sweep panel is set to measure the analyzer amplitude (Anlr
Ampl) and the source field is set to sweep the generator frequency (Gen Freq) which is swept
from 20 kHz to 20 Hz. The output should be set to V RMS and may be divided by the gain to get
the input referred noise voltage, though the data sheets normally specify the output noise
voltage in V RMS.
Audio Power Amplifier AP Analyzer In
CIN RFILT
IN+ OUT+ +
Diff Inputs
RL CFILT
CIN BTL RFILT Channel A
Outputs
IN- OUT-
-
CFILT Inputs Balanced

VS GND AC-Coupled
AP Generator Out Zin = 100 k / 185 pF
Internal Filter = 22 Hz - 22 kHz
+ RC Low-Pass Reading Meter = Amplitude
Filter for Data1 = Analyzer Amplitude
Channel A V+ GND Source = Generator Frequency
- Class-D
Measurements
Outputs Off (No Connect) Regulated

Sweep 20 kHz - 20 Hz Power Supply
Figure 8. Noise Measurement Circuit

SLOA068
RL = 8 AV = 6 dB Class-D
100 CB = 1 F BTL
VOUT (Vrms)
10
Class-AB
Class-D
1
20 200 2k 20k
Frequency (Hz)
Figure 9. Measured Results of Noise Circuit
5.2 Signal-to-Noise Ratio

The signal-to-noise ratio (SNR) is the measure of the maximum output voltage compared to the
integrated noise floor over the audio bandwidth, expressed in dB. It is normally specified at a
precise power in the data sheet tables. The integrated noise floor is measured using the
technique described in Section 5.1. The distortion of the output waveform is then measured at 1
kHz by sweeping the input voltage. The AP setup is the same as per the THD+N versus power
measurements, with VOUT, in V RMS, graphed on the x-axis rather than POUT. The point at which
the output voltage begins to clip (the THD+N increases sharply) is considered to be the
maximum output voltage.
The SNR is calculated using Equation (5). The noise and signal data can also be expressed in
decibel-volts (dBV), which is the dB ratio of the measured voltage to 1 V, and Equation (5) then
simplifies to Equation (6).
V RMS
SNR = 20 log OUT
(5)
V RMS
NOISE
SNR = dBVOUT dBVNOISE (6)
Any unused input should be ac-grounded. The measurement bandwidth should be limited to
provide an accurate measurement of the integrated noise floor.
6 Gain and Phase

The AP measurement circuit is shown in Figure 10 for a mono-channel, BTL-output APA.
Measurements for the TPA731 and TPA2001D1 are shown in Figures 11 and 12. The gain and
phase can also be measured at multiple points with an oscilloscope using Equation (7) for the
gain and Equation (8) for the phase, where t is the time delay between the input and output
voltages and f is the frequency of the input signal. The data is then plotted versus frequency.
V
A V (dB) = 20 log OUT (7)
VIN
= t f 360 o (8)

SLOA068
CIN RFILT
+ IN+ OUT+ +
Diff Inputs
RL CFILT
Channel A CIN BTL RFILT Channel A
Outputs
IN- OUT-
- -
CFILT
Outputs Balanced VS GND Inputs
CHA and CHB ON CHA Balanced, AC-Coupled
CHB Track CHA CHB Source set to GenMon
Zout = 40 Zin = 100 k /185 pF
Set Load Reference = RL Set Load Reference = RL
Sweep 20 kHz - 20 Hz Set dBrA Ref to Generator CHA
RC Low-Pass Internal Filter = <10 Hz - 80 kHz
V+ GND Filter for Class-D Reading Meter = Amplitude
Measurements Data1 = Analyzer Amplitude
Data2 = Analyzer Phase
Regulated Source1 = Generator
Power Supply Frequency
Figure 10. Gain and Phase Measurement Circuit
Figure 10 is the AP-II setup for measuring a single channel of the APA. Both channels must be
turned on at the generator panel in the software and CHB set to track CHA. The analyzer CHB
is set to GenMon (generator monitor), which means it takes its input directly from the generator
output of the selected channel internal to the AP-II and uses it as the input phase reference for
the analyzer measurement. The reference dBrA value should be set equal to the channel being
swept, which in this case is CHA. This sets the input voltage of channel A as the reference for
the gain measurement. It may be necessary to subtract 180 from the phase measurement to
get the actual phase value.
The APA input ac-coupling capacitors produce the phase shift and attenuation at low
frequencies. The class-D RC filter introduces some attenuation and phase shift at the
measurement endpoints as seen in Figure 12. The AP analyzer band-pass filters should be set
<10Hz and 30 kHz to minimize their impact on the measurement.
14 +40
Phase (Degrees)
12
Gain (dB)
10 VS = 3.3 V AV = 12 dB
RL = 8 PO = 250 mW Gain
CB = 1 F BTL Phase
-40
8 200 2k 20k
20
Frequency (Hz)
Figure 11. TPA731 Gain and Phase Measurements

SLOA068
+60
24
Phase (Degrees)
Gain (dB)
20
0
VS = 3.3 V AV = 23.5 dB
RL = 8 PO = 300 mW Gain
CB = 1 F BTL Phase
16
-40
20 200 2k 20k
Frequency (Hz)
Figure 12. TPA2001D1 Gain and Phase Measurements
7 Crosstalk
Crosstalk is the measure of the signal coupling between channels of a stereo device. The
crosstalk measurement circuit is shown in Figure 13 for an APA with differential inputs and BTL
outputs. This particular circuit is set up to measure right-to-left (R-L) channel crosstalk, or the
amount of signal that couples from the right channel (CHA) into the left channel (CHB). An input
signal is fed into the right channel and the outputs of both channels are measured and compared
as shown in Equation (9). The input voltage is fixed and is swept from 20 kHz to 20 Hz. The
setup is inverted to graph the L-R channel crosstalk and the terms in parentheses in Equation
(9) are inverted.
V OUT
Crosstalk = 20 log CHB
(9)
V OUT
CHA

SLOA068
CIN
+ IN+ OUT+ +
Right RL
Channel A CIN Channel A
Channel
-
IN- OUT- RC Low-Pass -
Filter for
CIN Class-D
+ IN+ OUT+ Measurements +
Left RL
Channel B CIN Channel Channel B
IN- OUT-
- -
Outputs Balanced VS GND Inputs

CHA and CHB ON Balanced, AC-Coupled
CHB Track CHA Zin = 100 k/185 pF
Zout = 40 Set Load Reference = RL
Set Load Reference = RL Set dBrB Ref to CHA
Sweep 20 kHz - 20 Hz Internal Filter = <10 Hz - 22 kHz
V+ GND Reading Meter = Crosstalk
Data1 = Analyzer Crosstalk
Source1 = Generator
Frequency
Regulated
Power Supply
Figure 13. Crosstalk Measurement Circuit
Both channels must be turned on at the generator panel in the software and CHB set to track
CHA. The input is swept over the audio frequency range at constant amplitude. The input
voltage should be set to the highest amplitude that does not cause the output voltage to clip.
Equation (10) is used for deriving the maximum peak-to-peak input voltage, where POUT(max) is the
maximum rated RMS output power, RL is the load resistance, and AV is the voltage gain of the
APA. The internal filter can be set to 30 kHz or greater to limit noise, but is otherwise not
required. The output cables of each channel should be separated to minimize capacitive
coupling between them.
2 2 POUT(max) R L
VIN(PP ) = (10)
AV
Connections for the measurements of SE devices are made in the same way as for BTL
devices, but with one end of RL tied to ground and a capacitor inserted between RL and OUT+ of
the APA. The measurement is taken across RL only, and not across RL and the capacitor.
A graph of the R-L crosstalk is shown in Figure 14. When both R-L and L-R crosstalk
measurements are shown, the graphs of both channels of the device are different. This is due to
impedance mismatch between the channels, which is caused by nonsymmetrical layout of the
IC.

SLOA068
The crosstalk was measured for the TPA0212 class-AB APA and TPA2001D2 class-D APA. The
values are in close agreement with the data sheet graphs. The class-D crosstalk improves as
the supply voltage is decreased because the radiation from the traces is decreased. Class-AB
amplifiers are relatively unaffected by changes in supply voltage. The crosstalk increases in all
amplifiers as the signal gain increases.
0
RL = 8 AV = 6 dB Class-D Class-AB
Crosstalk (dB)
CB = 1 F POUT = 250 mW Class-AB Class-D

-60 BTL POUT = 300 mW Class-D
-120
20 200 2k 20k
Frequency (Hz)
Figure 14. Crosstalk Measurements
8 Supply Rejection
Two types of supply rejection specifications exist: power supply rejection ratio (PSRR) and
supply ripple rejection ratio (kSVR). PSRR is a dc specification measuring the change in output
offset voltage for a change in supply voltage. kSVR is an ac specification measuring the ability of
the APA to reject ac-ripple voltage on the power supply bus. All power supply decoupling
capacitors are removed from class-AB circuits, and class-D measurements have a small 0.1F
decoupling capacitor placed close to the APA power pins to provide reverse path for recovery
switching currents. It is recommended that the designer use equal decoupling capacitance
values when comparing devices from different manufacturers to get a valid comparison of the
performance, because a higher capacitance equates to a better kSVR.
PSRR is the ratio of the change in the output voltage, VOUT(dc) for a change in the power supply
voltage, VS, expressed in dB as shown in Equation (11). For example, the output voltage of an
audio power amplifier that has a PSRR of -70 dB would change by 31.6V if the supply voltage
changed by 0.1V.
VOUT( dc )
PSRR = 20 log (11)
VS

kSVR is the ratio of the output ripple voltage, VOUT(ac), to the supply ripple voltage, expressed in dB
as shown in Equation (12). This parameter is normally listed as a typical value in the data sheet
tables at a specified frequency and temperature of 1 kHz and 25C, respectively. A graph is
provided in the data sheet of the typical values of kSVR over the audio bandwidth, because it is a
frequency-dependent parameter.
VOUT(ac )
k SVR = 20 log (12)
VS


SLOA068
The PSRR and kSVR measurement circuit is shown in Figure 15. The PSRR measurement
requires only the two DMMs; therefore RSVR, CSVR, the generator and analyzer, and the RC
measurement filter are not needed. The power supply voltage, VS, is initially set, then read from
the meter on the power supply. When the power supply meter does not have the desired
resolution, DMM1 is used to measure VS. DMM2 then measures VOUT across the load. VS is then
stepped up or down by a specific amount and the corresponding value of VOUT is measured.
The differences of the two measurements are then substituted into Equation (11) and the PSRR
is calculated for that specific change in supply voltage. PSRR is specified as a typical value that
is valid for a given supply voltage range at 25C. All APA inputs are ac-coupled to ground.
Audio Power Amplifier AP Analyzer In

VOUT
CIN (DMM2)
IN+ OUT+ RC Filter for +
Diff Input Filter-Free
BTL RL Channel B
CIN Class-D
Output Measurements
-
IN- OUT-
VS GND +
Channel A
-
C
AP Generator Out
CSVR
Inputs
+
VS Balanced, AC-Coupled
Channel A RSVR (DMM1) Zin = 100 k/185 pF
- Set Load Reference = RL
Internal Filter = <10 Hz - 80 kHz
Reading Meter = Crosstalk
Outputs Unbalanced-Float Data1 = Analyzer Crosstalk
V+ GND
CHA ON Source1 = Generator Frequency
CHB Track CHA
Zout = 20
Set Load Reference = RL Regulated
Sweep 20 kHz - 20 Hz Power Supply
The 0.1 F capacitor, C, is required for class-D operation.

The PSRR measurement uses the DMMs only because it is a dc value. kSVR measurements use either the analyzer, oscilloscope or DMMs
because it is an ac value. RSVR and CSVR are used for kSVR measurements only.
Figure 15. PSRR and kSVR Measurement Circuit
The kSVR measurement requires the generator, analyzer, a DMM, and the kSVR filter components
RSVR and CSVR. The RC measurement filter is used when the analyzer cannot accurately process
the square wave output of the filter-free class-D APAs. DMM1 is used to measure VS at the APA
power pins. The generator injects a small sine-wave signal onto the power bus, and the audio
analyzer measures this ac voltage at the APA power pin and at the output. Here the AP-II is
configured for a crosstalk measurement, and sweeps the ac voltage at constant amplitude over
the audio band, measuring and presenting a graph of the data points in dB.

SLOA068
Alternatives to the generator are to use a power source that has the capability to add an ac
component to the output, or use a transformer to couple the ac signal onto the power bus. In any
case, check the voltage that is applied to the APA power pins to be sure that the absolute
maximum ratings of the APA are not exceeded at any point during the process.
The kSVR filter circuit is shown in Figure 16. The dc power supply output impedance, RS, is
normally in the milli-ohms. The input impedance of the APA power pin, RAPA, is very high
compared to this (in the hundreds or the thousands). The generator output signal sees RAPA and
RS in parallel and, because of the low value of RS, this appears to be an ac ground. The resistor
RSVR is added to the circuit to increase the equivalent impedance of the power supply and is
chosen to be approximately equal to the output impedance of the ac signal generator, RGEN. A
voltage divider, formed between RSVR and RGEN, provides a reasonable amplitude ac signal at
the APA power pin. The large value of RSVR is tolerable because the dc and ac supply currents
are low. This is because the APA is idling and does not have any audio signal at the inputs, so
the power dissipated in RSVR is small.
RGEN CSVR RSVR
VGEN RAPA RS
Figure 16. kSVR Filter Circuit
The addition of CSVR ac-couples the generator to the power bus and provides a high-pass filter
for injecting the ac signal into the APA. The filter cutoff frequency, fC, should be set below the
lowest frequency of the audio band, fMIN, which in this case is 20 Hz. Equation (13) provides the
value for fC, which is ~14 Hz.
fMIN
fc = (13)
2
The equivalent resistance of Figure 20 is then calculated with Equation (14), where RAPA is the
supply voltage divided by the quiescent current of the device (VS/IQ). The value for CSVR is then
calculated using Equation (15).
R EQ = R GEN + R APA ll(R SVR + R S ) R GEN + R SVR (14)
1
C SVR = (15)
2 f C R EQ
The capacitor will most likely be electrolytic due to the value required. It will have some
reactance that will vary with frequency range as shown by Equation (16). At 20 Hz the
impedance will be quite highapproximately the value of RGEN and RSVRand at 20 kHz the
value will be in the milli-ohms.
1
X CSVR = (16)
2 f C C SVR

SLOA068
The actual values for the measurement circuit were RGEN = 20, RS = 0, RAPA = 5V/6mA = 833,
CSVR = 330F, RSVR = 20, fC = 12 Hz. This yields a capacitive reactance of 24 at 20 Hz, and
24 m at 20 kHz. The value of the ac signal may need to be adjusted at low frequencies so that
the desired voltage is applied to the APA power pin. The same is true for the dc voltage from the
power supply, since IQ will create a small voltage drop across RSVR.
Those devices with BYPASS pins will have improved kSVR as the capacitance on the pin is
increased. Devices operated SE have lower kSVR, particularly at the extreme low and high ranges
of the audio frequency band. This is primarily due to the large output ac coupling capacitor,
which dominates the frequency response both below and above the resonant frequency set by
the equivalent series resistance (ESR) and equivalent series inductance (ESL) of the capacitor.
The kSVR graphs are shown in Figure 17 for a 100-mV RMS input sine wave. Both of these
devices are differential input and BTL output. The TPA731 is measured with the inputs
floating, though newer devices are measured with the inputs ac-grounded. Figure 18 is a
data sheet graph from the TPA711 that provides an example of how CB impacts the kSVR
measurement of an SE output.
0
VS = 3.3 V AV = 12 dB Class-AB Class-AB
RL = 8 AV = 6 dB Class-D Class-D
KSVR (dB)
CB = 1 F BTL
-60
-100
20 200 2k 20k
Frequency (Hz)
Figure 17. kSVR of the TPA2001D1 and TPA731
0
VD D = 5 V
-10 RL = 8
SE
-20
C B = 0 .1 F
-30
-40
kSVR (dB)
-50
CB = 1F
-60
-70
-80
BYPASS = 1/2 VD D
-90
-100
20 100 1k 10k 20k
Frequency (Hz)
Figure 18. Impact of CBYPASS on kSVR for the TPA711 Class-AB APA

SLOA068
9 Power Measurements and Related Calculations
Several sets of data can be extracted from power measurements of a device. The power
measurement process begins with the primary measurement of amplifier efficiency. The power
that is dissipated by the amplifier is then calculated. This is useful for comparing the power
supply requirements of different devices. The crest factor (CF) of the audio signal directly
impacts the output power, and the effects are demonstrated from the dissipated power
calculations.
9.1 Efficiency Measurements

Efficiency is the measure of the amount of power that is delivered to a load for a given input
power provided by the supply. A class-AB APA acts like a variable resistor network between the
power supply and the load, with the output transistors operating in the linear region. They
dissipate quite a bit of power because of this mode of operation, and are therefore inefficient.
The output stage in class-D APA acts as a switch that has a small resistance when operated in
the saturation region, which provides a much higher efficiency.
A circuit for measuring the efficiency of a class-AB or class-D system is shown in Figure 19. The
simplest setup results when the power supply voltage and current meters have the resolution
required. When the supply current meter is not sufficient, R1 is placed in the circuit. It should be
a small value (0.1) and able to handle the power dissipated. A voltage drop occurs across R1,
so the supply voltage must be adjusted to set the desired VS at the device power pin. The
average voltage, V1, across R1 provides the average supply current (IS = V1/R1) that is used to
calculate the average power provided by the supply.
The true-RMS DMMs and the audio analyzer provide an RMS value of both the voltage and the
current, which, when multiplied together, provide the average power. When used, the power
supply meters provide the average value of the supply voltage and current. The oscilloscope can
measure the average or RMS values of the power supply and output voltage. Some
oscilloscopes even have current probes that can be used to measure the current through a wire,
in which case resistor R1 is not needed.
The load measurement is different for class-AB and class-D APAs. Two elements are shown;
one is the actual load, ZL, and the other is resistor R2. The Class-AB load is a noninductive
power resistor, ZL = RL, that must capable of handling the maximum power output without a
significant temperature increase, which will change the resistance and impact the measurement
accuracy. This purely resistive load makes the output measurement easy since only the voltage
across the load, VOUT, is required in order to calculate the output power. The output is sinusoidal
so all measurement devices should be ac-coupled to the load. There is some quiescent power
dissipation in RL, but this is negligible. Resistor R2 is not required for class-AB efficiency
measurements because the load is purely resistive.

SLOA068
The switching nature of the class-D makes the output measurement more challenging. First, a
speaker is used as the load for the filter-free class-D because it has the inductance that helps
provide the high class-D efficiency. A purely resistive load is not a true indicator of the operating
environment of the filter-free class-D, and does not provide accurate efficiency numbers.
Second, the output power must be calculated on the basis of current and voltage, not on the
basis of impedance, because impedance varies with frequency. A small power resistor (R2) is
placed in series with the load and a DMM or analyzer is used to measure the RMS value of the
load current (IOUT = V2/R2). The RMS voltage across the entire load (speaker and resistor R2)
must be measured to provide the total power into the load.
AP Generator Out Audio Power Amplifier Oscilloscope or Analyzer

V2
CIN
(DMM2) RC Filter
+ IN+ OUT+ +
Diff Input for
Channel A CIN BTL R2 Filter-Free VOUT Channel 1 (A)
Output Class-D -
IN- OUT-
-
Measurements
ZL
VS GND
Outputs Balanced V3
Zout = 40 (DMM3)
Set Frequency of Signal
+
R1 VS Channel 2 (B)
V1 -
(DMM1)
Inputs
Balanced, DC-Coupled
V+ GND
Zin = 100 k/185 pF
Set dBrB Ref to CHA
Regulated
Power Supply
Load ZL is a speaker for class-D APAs and is a purely resistive load for class-AB APAs
DMM1 and Channel 2 of the AP/oscilloscope (or a third DMM) are used to measure the average power supply
current and voltage when power supply meters are not accurate. If not used, remove resistor R1.
Figure 19. Efficiency Measurement Circuit for Class-AB and Class-D BTL APAs
Equation (17) provides the efficiency of the class-AB APA, and Equation (18) provides the
efficiency of the class-D APA. The input power of both equations, as stated previously, is just the
average voltage applied to the power pins of the APA multiplied by the average value of the
power supply current. Average value is used for the power supply measurements since the
voltage and current have dc and ac components and are typically nonsinusoidal. The output
power is also an average value that comes from the multiplication of two RMS terms.
VL ( RMS ) 2

P ZL
Class AB = OUT = (17)
V
PS S ( ave ) I S ( ave )

SLOA068
VR (RMS)
VO(RMS ) 2
POUT VO(RMS ) IO(RMS) R2

Class D = = = (18)
PS VS(ave ) IS(ave ) VS(ave ) IS(ave )
The RC measurement filter is used for making filter-free class-D output measurements when the
analyzer or DMM cannot accurately process the switching output waveform. The filter resistance
must be large enough to minimize current flow through the filter, while the capacitance must be
sized to achieve the desired cutoff frequency, which should be just above the audio band. If the
filter resistor is not large enough, the filter current must be accounted for in the efficiency
equation. The recommended values of RFILT and CFILT are 1 k and 5.6 nF, respectively. This
provides a filter cutoff frequency of ~28 kHz. The filter is only required with class-D APAs and is
discussed in more detail in Section 3.
The efficiency was measured with a 3.3-V supply and the results are shown in Table 3 and
Figure 20 using the power supply meter and a Fluke 87III DMM measuring the voltage across
the load. The DMM, AP analyzer, and TDS 754 oscilloscope measurements for the class-AB
data were in close agreement. The class-D DMM and AP data were similar, but the oscilloscope
measured 5-10% higher and is due to the averaging of the oscilloscope, which introduced a
somewhat large margin of error, particularly at high power output. The DMM reading is more
reliable since it filters out the high frequency harmonics of the switching waveform to provide a
more stable low-frequency value.
Table 3. Efficiency Data for the TPA731 and TPA2001D1
Vs Is Ps Vout Pout Eff Is Ps Vr Vout Pout Eff

(Vave) (mAave) (mWave) (mVrms) (mWave) (%) (mAave) (mWave) (mVrms) (mVrms) (mWave) (%)
3.3 23 75.9 200 5.0 6.6 3 9.9 0.7 58 0.4 4.1
3.3 28 92.4 250 7.8 8.5 4 13.2 1.3 104 1.4 10.2
3.3 40 132.0 354 15.7 11.9 5 16.5 2.3 200 4.6 27.9
3.3 45 148.5 400 20.0 13.5 8 26.4 3.7 335 12.4 47.0
3.3 56 184.8 500 31.3 16.9 10 33.0 4.5 393 17.7 53.6
3.3 67 221.1 600 45.0 20.4 13 42.9 5.1 486 24.8 57.8
3.3 79 260.7 708 62.7 24.0 17 56.1 6.3 594 37.4 66.7
3.3 89 293.7 798 79.6 27.1 22 72.6 7.4 688 50.9 70.1
3.3 111 366.3 998 124.5 34.0 29 95.7 8.8 824 72.5 75.8
3.3 134 442.2 1197 179.1 40.5 39 128.7 10.3 973 100.2 77.9
3.3 156 514.8 1397 244.0 47.4 55 181.5 12.7 1179 149.7 82.5
3.3 158 521.4 1417 251.0 48.1 74 244.2 15.0 1370 205.5 84.2
3.3 - - - - - 107 353.1 18.3 1664 304.5 86.2
3.3 - - - - - 144 475.2 21.2 1932 409.6 86.2

SLOA068
100
90
80
Efficiency (%)
70
60 Class-AB
50
40 Class-D
30
20
10
0
0 50 100 150 200 250 300 350 400 450
POUT (mW)
Figure 20. Efficiency Graphs of the TPA731 and TPA2001D1
9.2 Power Dissipated vs Power to the Load

The efficiency measurements provide the information required to calculate the amount of power
dissipated, PD, in the amplifier. PD provides some insight into the supply currents that are
required. PD is calculated using Equation (19) and the measured values of supply and output
power from Table 3. It is assumed that the power dissipated in the RC filter, used for the filter-
free class-D APA measurements, is negligible.
PD = PS POUT (19)
Figure 21 shows graphs of PD versus the POUT for the TPA731 class-AB and the TPA2001D1
filter-free class-D APAs, calculated from the efficiency data using Equation (19). The data was
measured up to the maximum output power, which occurs just prior to clipping, and can easily
be discerned from the THD vs Power graph. The designer can choose the percent distortion
(level of clipping) that is acceptable for a system and test the device through that power level.
300
250
200
Pd (mW)
Class-AB
150
Class-D
100
50
0
0 50 100 150 200 250 300 350 400 450
POUT (mW)
Figure 21. Graph of Power Dissipated vs Output Power

SLOA068
9.3 Crest Factor and Output Power
The crest factor (CF) is the ratio of the peak output to the average output. It is typically graphed
in terms of output power and is expressed in dB. For example, the CF of a sine wave is 3 dB.
Sine waves are used in the characterization of APA performance, but do not give a clear idea of
what the performance will be with music. The CF of music may vary between 6 dB and 24 dB.
The CF directly impacts the amount of heat dissipated in the device. The higher the CF, the
lower the heat dissipated and the higher the ambient operating temperature can be. The PD data
of Section 9.2 can be used to determine the CF of the device.
Equation (20) may be used to calculate CF. Since a sine wave was used for the measurements,
the CF is 3 dB, and the average output power (POUT(ave) ) is known. The peak output power
(POUT(pk) ) is calculated by manipulating Equation (20) into Equation (21), where POUT(pk) and
POUT(ave) are expressed in watts and CF is expressed in dB.
POUT(pk )
CF(dB) = 10 log (20)
POUT(ave )

POUT(pk )
POUT(ave ) = (21)
10 (CF 10 )
For example, the maximum peak output power is 500 mW at for the TPA731. This is calculated
using 250 mW as POUT(ave) and a CF of 3 dB for the output sinusoid. The peak will not change
throughout the calculations, as it is the maximum output power possible and is independent of
the output waveform. The CF is then increased in 3 dB steps up to 18 dB and the corresponding
POUT(ave) is calculated for each step. The PD in the device is measured for each value of POUT(ave)
using the efficiency measurement circuit.
The efficiency data and CF calculations can help the designer approximate the power that must
be provided by the power supply. Table 4 shows the values of power for the supply, load, and
what is dissipated in the amplifier for various CFs of the TPA731 class-AB APA and the
TPA2001D1 class-D APA. The table was generated from measured data and calculations using
Equations (19) through (21).
Figure 22 shows the graph of PS and POUT versus CF from the data of Table 4. The graph allows
easy comparison of the devices, and it is clear that the class-D APA provides more POUT with
less power from the supply than the class-AB APA. The difference between PS and POUT is the
dissipated power, PD.
Table 4. Power vs Crest Factor
Crest Crest
POUT Factor Ps Pd POUT Factor Ps Pd
(mWave) (dB) (mWave) (mWave) (mWave) (dB) (mWave) (mWave)
251 3 521 270 410 3 475 66
125 6 366 242 206 6 244 39
63 9 261 198 100 9 129 28
31 12 185 154 51 12 73 22
16 15 132 116 25 15 43 18
8 18 92 85 12 18 14 148

SLOA068
525
420 Po (Class-AB)
PS (mW) 315 Ps (Class-AB)
210 Po (Class-D)
105 Ps (Class-D)
0
3 6 9 12 15 18
Crest Factor (dB)
Figure 22. Supply and Output Power vs CF for the TPA731 and TPA2001D1
10 Measurement Pitfalls
This section contains a compilation of reminders to help avoid the various common mistakes, or
pitfalls, that are made when measuring the APA devices. While they are not all-inclusive, it is the
hope of the author that these may offer some insight that will save time and effort spent
troubleshooting the circuit.
10.1 Effects of Improper Interfacing and Grounding

The primary concern is establishing a good connection to the APA. A good connection allows
ground current to flow through a low-resistance return path and reduces noise injection into the
system through ground loops. Grounding is a critical part of this connection, particularly at the
APA inputs. THD+N levels were measured for various generator connections to a TPA2001D2
Class-D APA and are shown in Figure 23. The class-D has differential inputs and BTL outputs.
A balanced generator, used with differential inputs, has a maximum deviation of 0.02% THD+N
between a grounded and floating source at low power, a difference that is negligible. The
balanced generator provided the lowest value of distortion. It is comparable to an unbalanced
generator that has a floating source as long as the positive (+) and negative (-) pins of the
source are connected to the corresponding pins of the APA. The performance is degraded by
0.2% at lower power, and 0.01% at high power when the negative (-) pin is grounded at the
APA. If the generator source is grounded, the performance decreases by over 0.2% across the
power spectrum. A balanced source must therefore be used to remove the common-mode noise
and minimize offsets from ground currents to provide the most accurate measurement.

SLOA068
2.0
VS = 5 V AV = 6 dB
RL = 8 BTL
CB = 1 F
THD+N (%)
0.2
Balanced
Unbal-Flt
Unbal-Gnd
10m 100m 1
POUT (W)
Figure 23. Effect of Generator Interface on APA Measurements, THD+N vs Power Shown
It may be necessary to tie the ground pin of the power supply or other system device to chassis
ground to remove any 60-Hz component, called ac line or 60-Hz hum, from the signal path. This
must be done carefully or ground loops will be formed that will increase distortion. References 4
and 6 have more information on grounding and ground loops.
To sum up the APA connections:
Use a balanced source with differential inputs, unbalanced source with SE inputs.
Ground the power supply chassis to remove any 60-Hz hum.
The RC filter, used when measuring filter-free class-D APAs, should always be connected
to ground at the APA to allow a path for return currents and to minimize the ground loop
area.
The lead and/or wire lengths of the filter components should be kept as short as possible.
Power supply-to-APA and APA-to-load cables must be sized to avoid restricting the current
flow.
AC-ground all unused inputs during measurements.
Check to be sure the source is warmed up and all measurement devices are calibrated.
10.2 THD+N Measurements

The load resistance must be properly set in the analyzer software for correct output power.
In the case of high distortion at lower power, check the ground connections, generator
output configuration, and that the input and bypass capacitors are correct.
10.3 Noise Measurements

Limit the measurement to the audio band, because the noise value is integrated over the
specified frequency range.

SLOA068
10.4 Gain and Phase Measurements
Reference the output voltage to the input voltage.
Subtract 180 degrees from the phase when the phase shift is graphed greater than 180
degrees, which is often a characteristic of the analyzer.
Adjust the analyzer bandpass filters to less than 10 Hz and greater than 30 kHz to remove
their contribution to the phase shift in the audio band.
10.5 Crosstalk Measurements

The output cables of both channels should be twisted pair wires to minimize ground loops.
Reversed output connections result in a crosstalk that is measured in positive dB.
Unused APA inputs should be ac-coupled to ground. Floating inputs decrease crosstalk.
10.6 Supply Rejection Measurements

A 0.1F decoupling capacitor is required for class-D operation during these measurements.
All other capacitors should be removed. All decoupling capacitors should be removed for
class-AB measurements.
Be sure the output is being compared with the voltage at the power pins of the chip.
A small resistor (20 ) must be in series with the power supply to develop the input voltage.
As the value of bypass capacitance increases, kSVR improves (decreases).
10.7 Efficiency Measurements

Measure the supply voltage at the power pins of the chip.
The filter-free class-D RC measurement filter should have a high resistance for RFILT, with a
value of 1 k recommended. The current through the filter must be considered when the
value is smaller than this.
11 References
1. www.audioprecision.com, Audio Precision Website
2. Design Considerations for Class-D Audio Power Amplifiers (SLOA031)
3. Reducing and Eliminating the Class-D Output Filter (SLOA023)
4. Audio Measurement Handbook, Metzler, Bob, Audio Precision, 1993.
5. Introduction to Electroacoustics and Audio Amplifier Design, Leach, W. Marshall Jr.,
Kendall/Hunt Publishing, 1999
6. Noise Reduction Techniques in Electronic Systems; Ott, Henry W., Wiley Interscience, 1976

IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications,
enhancements, improvements, and other changes to its products and services at any time and to discontinue
any product or service without notice. Customers should obtain the latest relevant information before placing
orders and should verify that such information is current and complete. All products are sold subject to TIs terms
and conditions of sale supplied at the time of order acknowledgment.
TI warrants performance of its hardware products to the specifications applicable at the time of sale in
accordance with TIs standard warranty. Testing and other quality control techniques are used to the extent TI
deems necessary to support this warranty. Except where mandated by government requirements, testing of all
parameters of each product is not necessarily performed.
TI assumes no liability for applications assistance or customer product design. Customers are responsible for
their products and applications using TI components. To minimize the risks associated with customer products
and applications, customers should provide adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right,
copyright, mask work right, or other TI intellectual property right relating to any combination, machine, or process
in which TI products or services are used. Information published by TI regarding third-party products or services
does not constitute a license from TI to use such products or services or a warranty or endorsement thereof.
Use of such information may require a license from a third party under the patents or other intellectual property
of the third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of information in TI data books or data sheets is permissible only if reproduction is without
alteration and is accompanied by all associated warranties, conditions, limitations, and notices. Reproduction
of this information with alteration is an unfair and deceptive business practice. TI is not responsible or liable for
such altered documentation.
Resale of TI products or services with statements different from or beyond the parameters stated by TI for that
product or service voids all express and any implied warranties for the associated TI product or service and
is an unfair and deceptive business practice. TI is not responsible or liable for any such statements.
Following are URLs where you can obtain information on other Texas Instruments products and application
solutions:
Products Applications
Amplifiers amplifier.ti.com Audio www.ti.com/audio
Data Converters dataconverter.ti.com Automotive www.ti.com/automotive
DSP dsp.ti.com Broadband www.ti.com/broadband
Interface interface.ti.com Digital Control www.ti.com/digitalcontrol
Logic logic.ti.com Military www.ti.com/military
Power Mgmt power.ti.com Optical Networking www.ti.com/opticalnetwork
Microcontrollers microcontroller.ti.com Security www.ti.com/security
Telephony www.ti.com/telephony
Video & Imaging www.ti.com/video
Wireless www.ti.com/wireless
Mailing Address: Texas Instruments

Post Office Box 655303 Dallas, Texas 75265
Copyright 2004, Texas Instruments Incorporated

Application Report
SNAA050A November 2007 Revised May 2013
AN-1737 Managing EMI in Class D Audio Applications

.....................................................................................................................................................
ABSTRACT
Electromagnetic interference (EMI) is an unwanted disturbance caused in an electrical circuit by
electromagnetic radiation emitted from an external source. The disturbance may interrupt, obstruct, or
otherwise degrade the effective performance of the circuit.
Contents
1 EMI - The Basics ............................................................................................................ 2
2 The Class D Audio Amplifier .............................................................................................. 3
3 Countering EMI .............................................................................................................. 4
4 References ................................................................................................................... 6
List of Figures
1 Wavelength as a Function of Frequency ................................................................................ 2
2 Common Antennas ......................................................................................................... 2
3 Unintended Antennas in PCBs ............................................................................................ 3
4 The Class D Audio Amplifier PWM ....................................................................................... 4
5 Resistance and Inductance as a Function of Frequency .............................................................. 5
6 Comparison of Fixed-Frequency and Spread-Spectrum Modulation................................................. 5
7 LM48511 - A Spread-Spectrum Modulated Class D Amplifier ........................................................ 6
All trademarks are the property of their respective owners.
SNAA050A November 2007 Revised May 2013 AN-1737 Managing EMI in Class D Audio Applications 1
Submit Documentation Feedback
EMI - The Basics www.ti.com
1 EMI - The Basics

In todays portable and consumer applications, space has become a premium, and engineers are often
required to eliminate enclosures and shielding, and suppress EMI and noise through other means such as
better segregation at the circuit level. Smaller space and higher functionality require high-density PCBs,
and the use of wafer-scale packaging with tiny PCB design rules makes EMI more of a concern.
EMI encompasses two aspects. Emissions refer to the scope to which equipment generates radiated
noise. Susceptibility is the scope to which equipment is affected by emissions generated from other
electromagnetic waves. The degree to which a designer controls unintended emissions may make the
task of susceptibility easier. Emissions are generally classified as radiated and conducted emissions.
Radiated emissions leave a circuit board, trace, or wire, and propagate through the air in the form of
electromagnetic waves to interfere with a nearby receiver. It is important to note that a "receiver" refers to
any circuit whose operation can be affected adversely by the reception of electromagnetic energy - such
as a PCB trace or even the lead of an IC. Conducted emissions refer to energy which escapes, or is
conducted, out of a circuit through wires or cables. Conducted emissions may cause problems directly or
manifest themselves as radiated emissions.
In order to understand emissions, it is important to understand antennas. In Figure 1, the well-known
physical relationship between wavelength and frequency is shown:
Velocity of Light 300

O= =
Frequency x r
(MHz) x r
r is the relative permittivity
Figure 1. Wavelength as a Function of Frequency
The shortest length required to be an efficient antenna is /4. In the case of air, permittivity is 1, but in the
case of FR4 or glass-epoxy PCBs, permittivity is reduced to approximately 4.8. The effect causes a signal
traveling a trace to slow once it reaches the dielectric gradient created by the FR4 material, causing
essentially, a "wavelength- shortening" effect. For example, a 200 MHz signal has a quarter wavelength in
air of 16.7 cm.
In an inner-layer PCB trace, it is: 16.7/4.8(1/2) = 7.6 cm.
A PCB trace can act as an unintentional antenna even at lengths shorter than /4, increasing both
emissions and susceptibility. Surface traces also exhibit this wavelength shortening effect, as one side of
the dielectric serves to change the overall permittivity of the transmission.
Unintended antennas, such as PCB traces, are the key culprit behind radiated noise in digital systems. As
you will see, the Class D audio amplifier is, in essence, a digital system from the perspective of radiated
emissions. One key principle underlying electromagnetism is that of reciprocity. A flow of current can
create an electric field, and a change in electric flux can induce a current to flow. Likewise, an antenna
that is good for receiving is also good for transmitting. If any of the dimensions of an unintentional antenna
approach quarter wavelength at the frequency at which that antenna is excited by noise current, radiated
emissions can be expected.
y y
Figure 2. Common Antennas
2 AN-1737 Managing EMI in Class D Audio Applications SNAA050A November 2007 Revised May 2013
www.ti.com The Class D Audio Amplifier
These are two common antenna designs: the simple dipole and whip shown in Figure 2. An interesting
observation is that the whip is essentially one half of a dipole with the horizontal ground essentially being
induced to act as the other half of a dipole.
Antennas are intended to transmit and receive signals through the radiation of electric energy. But, as
illustrated in Figure 3, unintended antennas in circuit boards can include:
Long traces
Vias
Component leads and pins
Unpopulated PCB board connectors and headers
Figure 3. Unintended Antennas in PCBs
An unterminated surface trace or an unterminated buried trace can become an unintended whip antenna.
Segments of traces at different RF potential due to poor layout can become unintentional dipole antennas.
Also, the conductive layers of the PCB can act as the other leg of a dipole antenna with the plane itself
being coupled into the electric field.
2 The Class D Audio Amplifier

The Class D audio amplifier has emerged as a popular topology for the consumer market due to its high
efficiency. It modulates a high-frequency square wave by the incoming analog signal. The square wave
itself could be fixed frequency, variable frequency, or simply random pulses with no fixed frequency. A
low-pass filter, typically a 2-pole Butterworth, is used to filter the high-frequency content and recover the
original audio signal. In "filterless" topologies, the inductance of the speaker itself is incorporated as part of
the filter. One common Class D topology, pulse width modulation (PWM), uses a fixed-frequency
waveform and changes the duty cycle to create a moving average of the signal after a low-pass filter as
seen in Figure 4.
The benefits of a switching topology are apparent: high efficiency, low power consumption, and small
thermal designs. But increased efficiency is not without cost. In order to drive efficiencies up, a sharp,
rapidly switching square wave is required. This can lead to the same undesirable artifacts that are present
in digital systems as the spectral energy is highly concentrated on the edges of the square wave. Also,
there will be some overshoot, causing the waveform to go beyond the maximum and minimum voltages
for a short time. Overshoot creates additional high-frequency content in the output spectrum and is
undesirable for EMI and audio performance.
Countering EMI www.ti.com
+VCC
+
Q1 IN
VIN -
LF
CF
IP IL
Q2
-VEE
Figure 4. The Class D Audio Amplifier PWM
3 Countering EMI
To counter EMI, it is essential to include PCB floor planning as part of the circuit design with the electrical
engineer, PCB layout engineer, and manufacturing engineer. General PCB guidelines for dealing with EMI
include:
Placement of decoupling capacitors between power and ground where voltage fluctuations are
determined to exist. Haphazard placement of caps can exacerbate the EMI problem. References for
the use of decoupling capacitors are included at the end of this application report
Power planes should be backed off from the edges of the PCB
Avoidance of traces cut within ground or power planes, which can create an unintended aperture
Adequate termination of all high-frequency clock lines.
Proper filtering of PCB connectors
Good PCB design will avoid loop antennas. Loop antennas encompass any route in which both
forward and return currents are on a well-defined conducting path.
Optimally, you will stop radiation by suppressing the source of current that is feeding the antenna.
For the audio designer, it is important to consider the following:
Keep traces from the audio amplifier to the speaker as short as possible. PCB traces and wires act as
antennas with significant radiation occurring once the trace length reaches /4.
For filterless Class D systems, the trace and cable length connecting the amplifiers output to the
speakers will likely be the largest source of RF emissions.
The practice of placing ferrite beads in series with the loudspeakers close to the amplifier can be effective.
In order to better understand the suppression mechanism of ferrite beads, it is useful to break the ferrite
bead down into frequency-dependent resistive and inductive elements - R(f ) and L(f ) - as shown in
Figure 5.
Regarding EMI suppression purposes, ferrite beads act as resistors, but because RDC = 0, there is no DC
voltage drop. This makes them useful in cases where frequencies of interest are significantly below 1
MHz. Also, as shown in Figure 5, it is important to understand that the ferrite bead is effective when
considered part of a two-element voltage divider. Both Z1 and Z2 are frequency dependent. To achieve the
desired function of a low-pass filter, the following relationships should be present: Z2 >Z1 at desired
frequencies and Z1>Z2 at noise frequencies.
www.ti.com Countering EMI
Z1
R(f) L(f)
Z2
Figure 5. Resistance and Inductance as a Function of Frequency
The ferrite usually serves as the series element, and the shunt element is a capacitance - either a physical
capacitance or a lumped capacitance. The transfer function indicates that Z1 increases and Z2 decreases
with frequency (1/jC) and that the system is damped to the extent that resonance effects are significantly
diminished. The primary difficulty with periodic square waves inherent in Class D is the concentration of
energy at the harmonic intervals.
In the push to create a "quiet" low-EMI class D amplifier, one approach is to dither that frequency back
and forth or spread the spectrum of the switching so the energy at any one point in the spectrum is
reduced. Compared to traditional Class D amplifiers, the spread-spectrum-modulation scheme has some
key advantages. Efficiency and low THD+N are maintained, but radiated noise and EMI can be
significantly reduced, as seen in Figure 6.
Figure 6. Comparison of Fixed-Frequency and Spread-Spectrum Modulation
The LM48511 is a spread-spectrum-modulated class D audio amplifier. It also includes a built-in boost
regulator that drives the supply voltage to 7 V, increasing amplifier output power and the audio-sound
pressure level compared to an unboosted amplifier. The boost regulator allows the amplifier to maintain a
constant output level, even when powered from a decaying voltage source such as a battery.
The LM48511 amplifier features a logic-selectable, spread-spectrum modulator that reduces EMI,
eliminating the need for output filters or chokes. As shown in Figure 7, the spread-spectrum modulator
feeds a standard H-bridge that drives the bridgetied-load speaker. In spread-spectrum mode, the
switching frequency varies randomly by 10% around a 330 kHz center frequency, decreasing EMI
emissions radiated by the speaker and associated cables and traces. Where a fixed-frequency Class D
exhibits large amounts of spectral energy at multiples of the switching frequency, the spread-spectrum
architecture of the LM48511 amplifier spreads that energy over a larger bandwidth, reducing peak noise
power in the circuit.
Electromagnetic interference is a system-level concern, and it is essential for todays audio engineer to
design with EMI in mind, including the best possible design practices and judicious choice of components
and materials.
References www.ti.com
+3.0V to + 5.5V
L1
6.8 PH D1
CS1
10 PF
VDD C2
SW
100 PF
C1
280 pF
R3
25.5 k:
R4
REGGND 2.5 k:
SD_BOOST
MODULATOR
FB
SOFTSTART
CSS R2 R1
0.1 PF 9.31 k: 4.87 k:
FB_GND1
OSCILLATOR
FB_SEL
FB_GND0
V1
SD_AMP
PV1
C4 C3
VGO- 1 PF 1 PF
R6
CIN R5
20 k:
20 k:
IN+
VIN+ + LS+
R7 H-BRIDGE
MODULATOR LS-
20 k: IN-
VIN- -
CIN R8
20 k:
VGO+
OSCILLATOR
SS/FF
GND LSGND
Figure 7. LM48511 - A Spread-Spectrum Modulated Class D Amplifier
4 References
Bruce Archambeault and James Drewniak, "PCB Design for Real-World EMI Control", 2002
David Terrell and R. Kenneth Keenen, "Digital Design for Interference Specifications", 1997
Howard Johnson, "High-Speed Digital Design: A Handbook of Black Magic", 1993
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, enhancements, improvements and other
changes to its semiconductor products and services per JESD46, latest issue, and to discontinue any product or service per JESD48, latest
issue. Buyers should obtain the latest relevant information before placing orders and should verify that such information is current and
complete. All semiconductor products (also referred to herein as components) are sold subject to TIs terms and conditions of sale
supplied at the time of order acknowledgment.
TI warrants performance of its components to the specifications applicable at the time of sale, in accordance with the warranty in TIs terms
and conditions of sale of semiconductor products. Testing and other quality control techniques are used to the extent TI deems necessary
to support this warranty. Except where mandated by applicable law, testing of all parameters of each component is not necessarily
performed.
TI assumes no liability for applications assistance or the design of Buyers products. Buyers are responsible for their products and
applications using TI components. To minimize the risks associated with Buyers products and applications, Buyers should provide
adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any patent right, copyright, mask work right, or
other intellectual property right relating to any combination, machine, or process in which TI components or services are used. Information
published by TI regarding third-party products or services does not constitute a license to use such products or services or a warranty or
endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property of the
third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of significant portions of TI information in TI data books or data sheets is permissible only if reproduction is without alteration
and is accompanied by all associated warranties, conditions, limitations, and notices. TI is not responsible or liable for such altered
documentation. Information of third parties may be subject to additional restrictions.
Resale of TI components or services with statements different from or beyond the parameters stated by TI for that component or service
voids all express and any implied warranties for the associated TI component or service and is an unfair and deceptive business practice.
TI is not responsible or liable for any such statements.
Buyer acknowledges and agrees that it is solely responsible for compliance with all legal, regulatory and safety-related requirements
concerning its products, and any use of TI components in its applications, notwithstanding any applications-related information or support
that may be provided by TI. Buyer represents and agrees that it has all the necessary expertise to create and implement safeguards which
anticipate dangerous consequences of failures, monitor failures and their consequences, lessen the likelihood of failures that might cause
harm and take appropriate remedial actions. Buyer will fully indemnify TI and its representatives against any damages arising out of the use
of any TI components in safety-critical applications.
In some cases, TI components may be promoted specifically to facilitate safety-related applications. With such components, TIs goal is to
help enable customers to design and create their own end-product solutions that meet applicable functional safety standards and
requirements. Nonetheless, such components are subject to these terms.
No TI components are authorized for use in FDA Class III (or similar life-critical medical equipment) unless authorized officers of the parties
have executed a special agreement specifically governing such use.
Only those TI components which TI has specifically designated as military grade or enhanced plastic are designed and intended for use in
military/aerospace applications or environments. Buyer acknowledges and agrees that any military or aerospace use of TI components
which have not been so designated is solely at the Buyer's risk, and that Buyer is solely responsible for compliance with all legal and
regulatory requirements in connection with such use.
TI has specifically designated certain components as meeting ISO/TS16949 requirements, mainly for automotive use. In any case of use of
non-designated products, TI will not be responsible for any failure to meet ISO/TS16949.
Audio www.ti.com/audio Automotive and Transportation www.ti.com/automotive
Amplifiers amplifier.ti.com Communications and Telecom www.ti.com/communications
Data Converters dataconverter.ti.com Computers and Peripherals www.ti.com/computers
DLP Products www.dlp.com Consumer Electronics www.ti.com/consumer-apps
DSP dsp.ti.com Energy and Lighting www.ti.com/energy
Clocks and Timers www.ti.com/clocks Industrial www.ti.com/industrial
Interface interface.ti.com Medical www.ti.com/medical
Logic logic.ti.com Security www.ti.com/security
Power Mgmt power.ti.com Space, Avionics and Defense www.ti.com/space-avionics-defense
Microcontrollers microcontroller.ti.com Video and Imaging www.ti.com/video
RFID www.ti-rfid.com
OMAP Applications Processors www.ti.com/omap TI E2E Community e2e.ti.com
Wireless Connectivity www.ti.com/wirelessconnectivity
Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265
Application Report
SNAA057B June 2008 Revised May 2013
AN-1849 An Audio Amplifier Power Supply Design

.....................................................................................................................................................
ABSTRACT
This application report provides design information for a power supply for use with our newest offering of
high-performance, ultra high-fidelity audio amplifier input stage ICs.
Contents
1 Introduction .................................................................................................................. 2
2 Overview ..................................................................................................................... 2
3 Schematic and Design ..................................................................................................... 3
3.1 Power Supply ....................................................................................................... 3
4 Bill Of Materials ............................................................................................................. 4
5 Additional Circuit ............................................................................................................ 6
5.1 120V/240V Selection Option ..................................................................................... 6
5.2 Inrush Current Control ............................................................................................. 7
5.3 Power Up/Down Mute Control .................................................................................... 8
6 Summary ................................................................................................................... 11
7 Board Layer Views ........................................................................................................ 12
8 Revision History ........................................................................................................... 16
List of Figures
1 Complete Power Supply Circuit ........................................................................................... 4
2 120V Transformer Connections, Primaries in Parallel ................................................................. 6
3 120V Transformer Connections, Primaries in Series................................................................... 6
4 Inrush Current Control ...................................................................................................... 7
5 Supply Ramp at Power On ................................................................................................ 7
6 Mute Control ................................................................................................................. 8
7 Mute at Power On........................................................................................................... 9
8 Mute at Power Off........................................................................................................... 9
9 Constant Brightness LED Circuit ........................................................................................ 10
10 Constant Brightness LED and Mute Control Circuit .................................................................. 10
11 PCB Composite View From Top ........................................................................................ 12
12 PCB Top Silkscreen View ................................................................................................ 13
13 PCB Bottom Silkscreen View ............................................................................................ 14
14 PCB Top Layer View ...................................................................................................... 15
15 PCB Bottom Layer View .................................................................................................. 16
List of Tables
1 Bill Of Materials ............................................................................................................. 4
All trademarks are the property of their respective owners.
SNAA057B June 2008 Revised May 2013 AN-1849 An Audio Amplifier Power Supply Design 1
Introduction www.ti.com
1 Introduction
Analog audio circuit power supplies can have an audible effect in listening test and quantifiable effect in
bench measurement results. Power supply designs that operate from the power mains are of three
common types: Switch mode (SMPS), regulated, and unregulated power supplies.
Switch mode power supplies have become very popular, common, inexpensive, and readily available.
SMPS are used extensively in computer hardware. They are well suited for such use providing good
regulation with high efficiency in a small physical size. A drawback to SMPS is the switching nature of the
design which creates EMI and RFI plus electrical noise on the supply rails. Small signal analog circuits are
more susceptible to noise in the form of EMI or electrical noise on the supply lines. Certain classes of
amplifiers, namely Class G and Class H, may be more easily realized with SMPS that are fast responding
for full audio bandwidth signals. Using SMPS for audio circuits presents additional design challenges than
when using a SMPS for non-audio circuits.
A regulated supply can be a simple linear regulator IC with the rectified voltage from the transformer as
input and a handful of external components or any number of more complicated and often higher
performance designs. There are the tradeoffs of complexity, cost, space, thermal design, reliability and
protection with any regulated design. It is common for regulated supplies to be used for the analog small
signal portions and other sensitive circuits for best performance. For an audio power amplifier, regulated
supplies will need high bandwidth for good audio performance. The complexity and cost for such a power
supply design may not be acceptable. Most linear regulator ICs do not have high bandwidth and are slow
compared to audio signals that can result in reduced audio performance.
For simplicity, good performance, and reasonable cost, an unregulated supply is the most common for an
audio power amplifier. An unregulated supply uses a transformer, a bridge rectifier, and various rail
capacitors. A draw back to the unregulated supply is the voltage fluctuations with load and power mains
fluctuations. A design should allow for a minimum 10% high line condition on the power mains.
Unregulated supplies may have only a fuse in the power mains input to protect against excessive current
unlike more sophisticated regulated designs. Additionally, the power supply voltage rails may have inline
fuses to add some additional protection.
The circuit and solution presented in this application note has not been tested to any industry standards. It
is the responsibility of the reader to perform standard industry testing to assure safety when using the
solution in part or in whole in any form. Texas Instruments does not provide any guarantees, written or
implied, about the safety of the solution.
2 Overview
This application note will cover the design of a 72V unregulated power supply designed specifically for
the LME49810, LME49811 and LME49830 high-fidelity audio amplifier modules. The output power of the
modules are approximately 220W to 250W into 8 and 350W to 400W into 4. Complete documentation
for the amplifier modules can be found in the LME49830TB Ultra-High Fidelity High Power Amplifier
Reference Design (SNAA058).
Although the power supply design is specific to the amplifier modules the concepts and circuit design may
be used for any power supply purpose.
The power supply is an unregulated design with an option to allow connection to either 120V or 240V
mains. The design uses toroidal transformers, a fully integrated bridge, and various rail capacitors for
ripple voltage reduction, noise suppression, and to act as high current reservoirs. Additional circuitry to
control inrush current on power up and power up/down Mute control are also included. A complete
schematic, PCB views, and Bill of Materials are provided for the power supply design.
2 AN-1849 An Audio Amplifier Power Supply Design SNAA057B June 2008 Revised May 2013
www.ti.com Schematic and Design
3 Schematic and Design
3.1 Power Supply

Figure 1 shows the complete schematic of the power supply design. The heart of the design is the basic
power supply consisting of the transformers, the bridge, and various capacitors. Many of the capacitors
used may not be commercially necessary or may have a minimal effect on performance. Because the
design is not a commercial design where tight cost constraints must be taken into account, additional
capacitors are freely used. For a commercial design, bench and listening test or some other test criteria is
recommended to determine the exact number, size, and type of external components required. A short
explanation of the purpose of each capacitor at the primary side of the transformers, around the bridge
and on the supply rails follows. Some capacitors are doubled up on the PCB for flexibility or to achieve the
desired total capacitance.
C1, C2, C4 are to protect against turn on/off spikes caused when the power switch changes positions.
C3 is not used and is redundant.
CS1, CS2 are low value, ceramic capacitors to filter higher frequency noise right at the DC output of the
diode bridge.
CS3, CS4 are the large reservoir capacitors to supply large current demands and stabilize the supply
rails to minimize low frequency fluctuations. These are very large value electrolytic capacitors. Two
capacitors are used to achieve the desired 40,000F capacitance per rail.
CS5, CS6 are high quality film capacitors to filter higher frequency noise. Two footprints are used on the
PCB for flexibility.
CS7, CS8 act in conjunction with RS1 and RS2 to decouple the large electrolytic capacitors and reduce
impedance.
CS9, CS10 are low value, ceramic capacitors to filter higher frequency noise from the transformer
secondary AC lines at the diode bridge.
CS11 - CS14 are in parallel with the bridge diodes to reduce high frequency noise and ringing of the
diode. An additional RC snubber in parallel with each diode of the rectifier will further reduce noise and
ringing.
The values for the different capacitors were not chosen based on extensive bench work or research. The
values were chosen based on general guidelines and commonly used values. Additional performance may
be obtained through refinement of the capacitor values. The equations and methods to determine optimal
values are beyond the scope of this application note.
Additionally, the supply rails have bleeder resistors, RBL1, RBL2, to drain the large reservoir capacitors (CS3,
CS4). Two footprints per rail were placed on the PCB to allow for lower power resistors to be used and a
wide range of bleeder current. More sophistication can be added by including an additional DPDT relay
and controls to only connect the bleeder resistors below a set voltage and remain unconnected during
normal operation.
The fully integrated bridge has a peel and stick heat sink attached. See Table 1 for robustness in use and
higher ambient temperature conditions.
Bill Of Materials www.ti.com
Figure 1. Complete Power Supply Circuit
4 Bill Of Materials
Table 1. Bill Of Materials

Reference Value Tolerance Description Manufacturer Part Number
400V, metalized
C1, C2, C4 0.01F 10% polyester film, Panasonic ECQ-E4103KF
7.5mm lead spacing
C3 Not Used
100V ceramic, X7R
CS1, CS2, CS7,
0.1F 10% type, 200mil lead AVX Corporation SR211C104KAR
CS8, CS9, CS10,
spacing
250V, metalized
CS11, CS12, CS13,
0.1F 10% polyester film, Panasonic ECQ-E2104KF
CS14
7.5mm lead spacing
CS3A, CS3B, 100V electrolytic CDE Cornell DCMC203U100BC2
20,000F 20%
CS4A, CS4B can Dubilier B
100V, metalized
CS5A, CS5B,
1F 10% polyester film, Panasonic ECQ-E1105KF
CS6A, CS6B
10mm lead spacing
www.ti.com Bill Of Materials
Table 1. Bill Of Materials (continued)

Reference Value Tolerance Description Manufacturer Part Number
63V electrolytic
CSR1, CSR2 1F 20% radial, 2mm lead Panasonic EEU-EB1J1R0S
spacing
Vishay
D1 1A 400V diode, DO-41 1N4004-E3/54
Semiconductor
2W Zener diode, Microsemi
DZ1 51V 5% 2EZ51D5DO41
DO-41 Corporation
2W Zener diode, Microsemi
DZ2 43V 5% 2EZ43D5DO41
DO-41 Corporation
500mW Zener
DZM 3.9V 5% Diodes Inc. 1N5228B-T
diode, DO-35
RBLD1, RBLD2, International Yageo
2k 5% 5W metal oxide SQP500JB-2K0
RBLD3, RBLD4 Corporation
International Yageo
RFAN 1.2k 5% 5W metal oxide SQP500JB-1K2
Corporation
5W wirewound Huntington Electric,
RIR1, RIR2, RIR3 68 1% ALSR-5-68-1%
silicone Inc.
RS1, RS2 1 5% Watt carbon film Panasonic ERD-S2TJ1R0V
International Yageo
RG 100 1% Watt metal film MFR-25FBF-100R
Corporation
1 Watt metal oxide
RZ1 560 5% Panasonic ERG-1SJ561
film
RZ2 390 5% Watt carbon film Panasonic ERD-S1TJ391V
RPD 10k 5% Watt carbon film Panasonic ERD-S2TJ103V
48V, 400mW SPST, Panasonic Electric
RL1 16A ALE15B48
N.O., relay Works
Fairchild
U1 35A 700V bridge rectifier GBPC3510W
Semiconductor
DPDT PCB mount,
S1 6A C&K Components 1201M2S1CQE2
mini slide switch
3 pin 156mil
Molex/Waldom
J1, J5 header, right angle, 26-60-5030
Electronics Corp.
tin plating
2 pin 156mil
Molex/Waldom
J2, J9, J4A, J4B header, right angle, 26-60-5020
Electronics Corp.
tin plating
4 pin 156mil
Molex/Waldom
J3A, J3B header, right angle, 26-60-5040
Electronics Corp.
tin plating
2 pin 100mil
J7, J8, J11, J12, Molex/Waldom
header, right angle, 22-05-3021
J13, J14, J15 Electronics Corp.
tin plating
Dual primary, dual
Transformer1, Plitron
24V, 300VA secondary, torrid 77060201
Transformer2 Manufacturing Inc.
transformer
Peel & stick heat
sink for bridge, CTS Electronic
CA = 16.5C/W BDN12-5CB/A01
1.21" square, 0.55" Components, Inc
tall
RZ3, RZ4, DZ3,
Option unused
DZ4, CSF1, CSF2,
circuits
CSF3
Additional Circuit www.ti.com
5 Additional Circuit
5.1 120V/240V Selection Option

For multi-country operation a switch is included to select between 120V or 240V input at the primary side
of the transformers. The transformers are dual primary with the switch allowing the option to put the
primaries into series or parallel. The primary side of each transformer is connected in parallel for 120V
operation with series connection used for 240V operation. The schematics, Figure 2 and Figure 3, show
the different connections with the switch set for either 120V or 240V input from the power lines.
Figure 2. 120V Transformer Connections, Primaries in Parallel
Figure 3. 120V Transformer Connections, Primaries in Series
www.ti.com Additional Circuit
5.2 Inrush Current Control

A simple inrush circuit is used to limit the high current that occurs at power up. The portion of the
schematic that controls inrush current is shown in Figure 4.
Figure 4. Inrush Current Control
The inrush circuit consist of three 68/5W resistors (RIR1 - RIR3, labeled just RIR in Figure 1 and Figure 4)
in parallel, a relay and the relay controls. The RIR resistors limit transformer primary current flow and the
resulting secondary current flow when the transformer is powered for a softer turn on. Once the VCC rail
voltage exceeds 33V the relay is activated shorting out the resistors. The relay is deactivated when the
VCC voltage falls below 10V resetting the circuit. The circuit is very simple and does not limit inrush current
if the mains power is switched on before the VCC rail drops below 10V. The relay control consists of the RZ1
and RZ2 resistors to limit current through the voltage clamping DZ2 Zener diode. DZ2 limits the relay voltage
below the maximum 48V rating. The D1 diode is for the relay coil EMF and CSR2 is to remove ripple and
stabilize the relay voltage. The oscilloscope view in Figure 5 shows how the positive rail charges up with
the increase in charge rate once the relay is closed shorting out the inrush current limiting resistors. The
RIR resistors will get warm but they are only conducting for 500ms each time the amplifier is powered on
keeping the power dissipation well within the 5W rating.
Figure 5. Supply Ramp at Power On
5.3 Power Up/Down Mute Control

The Mute function of the audio amplifier input stage IC is used for a completely quiet turn on and turn off.
The amplifier is held in Mute mode until the voltage supplies are nearly stable and also goes into Mute
mode once the supplies have collapsed below a determined voltage. With 40,000F of supply reservoir
capacitance per rail the amplifier can continue operation for some time after the mains power has been
removed. The mute control circuit removes the drive signal for a quicker turn off well before the supplies
have collapsed down below the minimal operating voltages. The amplifier will turn off quietly and smoothly
without any undesired noise. The Mute control circuit portion is shown in Figure 6.
Figure 6. Mute Control
The voltage threshold is set by the value of the DZ1 Zener diode, the current limiting RZ1 resistor and the
forward voltage on the LED. The circuit works by simply requiring a certain positive supply rail voltage
before the LED turns on and the amplifier switches out of Mute mode. The DZ1 Zener diode will begin to
conduct once the positive supply rail exceeds it's rated voltage. At this point the LED will begin to develop
voltage across it. The LED's forward voltage (typically 2V ~ 4V) is used as the amplifier's Mute voltage.
Setting the Mute resistor on the amplifier PCB module correctly allows the amplifier to go out of Mute
mode once the LED's forward voltage is high enough to supply the needed Mute current. The LED is also
used as an indicator, lighting when the amplifier is in Play mode. The values shown set the Mute voltage
threshold to 57V on power up and 58V on power down. Because of component tolerances the threshold
voltages will vary. At power down, the forward voltage of the LED will collapse quickly putting the amplifier
into Mute mode well before the supplies are discharged for a quiet and relatively quick power off. Figure 7
and Figure 8 show the Mute signal with supply voltage at power on and power off. There is additional
delay from when the Mute signal reaches the Mute threshold (~1.80V for the amplifier PCB) and when the
amplifier enters PLAY mode as a result of the mute delay capacitor on the amplifier PCB.
www.ti.com Additional Circuit
Figure 7. Mute at Power On
Figure 8. Mute at Power Off
The RZM Zener diode is for protection in the event of LED failure locking the Mute voltage so it will not
exceed 4V. The amplifier PCB module's Mute resistor is sized for a maximum of 4V safely limiting Mute
current. RPD is needed so DZ1 will conduct and CSR1 is for a steady LED/Mute voltage.
A short coming of the simple Mute control circuit is the LED's brightness will vary under heavy amplifier
load with the circuit values shown in Figure 6. Either the threshold of the Mute circuit can be lowered by
changing the value of DZ1 for more consistent brightness in operation or a constant current circuit may be
used. Figure 9 shows a basic constant current (LED brightness) circuit with similar threshold voltages as
the Mute control circuit.
Figure 9. Constant Brightness LED Circuit
The LED will first begin to light when the positive supply rail voltage exceeds 45V. Once the positive rail
reaches 60V the LED will have 6.5mA of current and only increase to 6.7mA at 80V with indiscernible
change in brightness. Zener diode DZA sets the minimum threshold for first light of the LED. Combining the
values of DZA, DZB, along with voltage drop across R1 sets the voltage when the LED current reaches a
constant value and constant brightness. R3 and DZC set the LED current and R2 is used to bias QLED and
limit current through DZC. By using a 10V Zener diode (DZB) the power dissipation in QLED is kept very low
so that a small transistor can be used without power dissipation concerns. The trade-off is that the DZA
Zener diode is required to dissipation about 1W when the supply reaches 80V. Figure 9 does not give
both constant LED current and the Mute signal control as Figure 6, although the Mute control could be
taken at the emitter of QLED. An alternate circuit to combine both Figure 6 and Figure 9 is shown in
Figure 10.
Figure 10. Constant Brightness LED and Mute Control Circuit
www.ti.com Summary
The circuit in Figure 10 will have the same threshold voltages as in Figure 9 and similar Mute control
thresholds as in Figure 6 but can also be used to control the Mute signal to the audio amplifier module.
For a reduced supply voltage window from LED first light to constant brightness, DZA should be increased
while DZB is reduced. This will increase the LED first light threshold while reducing the additional voltage
needed to reach the constant brightness threshold. The value of DZC may also be adjusted to achieve the
designed circuit response.
6 Summary
The unregulated power supply presented will give very good performance while powering an audio
amplifier. While circuit modifications and additions can improve performance the solution presented has a
relatively low part count and simplicity is maintained with all circuits. The power supply will provide a 70V
to 73V supply under quiescent conditions with full load voltage dropping to 59V to 62V.
Board Layer Views www.ti.com
7 Board Layer Views
Figure 11. PCB Composite View From Top
www.ti.com Board Layer Views
Figure 12. PCB Top Silkscreen View
Board Layer Views www.ti.com
Figure 13. PCB Bottom Silkscreen View
www.ti.com Board Layer Views
Figure 14. PCB Top Layer View
Revision History www.ti.com
Figure 15. PCB Bottom Layer View
8 Revision History
Rev Date Description

1.0 06/03/08 Initial release.
1.01 03/15/10 Deleted all references to AN-1625.
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, enhancements, improvements and other
changes to its semiconductor products and services per JESD46, latest issue, and to discontinue any product or service per JESD48, latest
issue. Buyers should obtain the latest relevant information before placing orders and should verify that such information is current and
complete. All semiconductor products (also referred to herein as components) are sold subject to TIs terms and conditions of sale
supplied at the time of order acknowledgment.
TI warrants performance of its components to the specifications applicable at the time of sale, in accordance with the warranty in TIs terms
and conditions of sale of semiconductor products. Testing and other quality control techniques are used to the extent TI deems necessary
to support this warranty. Except where mandated by applicable law, testing of all parameters of each component is not necessarily
performed.
TI assumes no liability for applications assistance or the design of Buyers products. Buyers are responsible for their products and
applications using TI components. To minimize the risks associated with Buyers products and applications, Buyers should provide
TI does not warrant or represent that any license, either express or implied, is granted under any patent right, copyright, mask work right, or
other intellectual property right relating to any combination, machine, or process in which TI components or services are used. Information
published by TI regarding third-party products or services does not constitute a license to use such products or services or a warranty or
endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property of the
third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of significant portions of TI information in TI data books or data sheets is permissible only if reproduction is without alteration
and is accompanied by all associated warranties, conditions, limitations, and notices. TI is not responsible or liable for such altered
documentation. Information of third parties may be subject to additional restrictions.
Resale of TI components or services with statements different from or beyond the parameters stated by TI for that component or service
voids all express and any implied warranties for the associated TI component or service and is an unfair and deceptive business practice.
TI is not responsible or liable for any such statements.
Buyer acknowledges and agrees that it is solely responsible for compliance with all legal, regulatory and safety-related requirements
concerning its products, and any use of TI components in its applications, notwithstanding any applications-related information or support
that may be provided by TI. Buyer represents and agrees that it has all the necessary expertise to create and implement safeguards which
anticipate dangerous consequences of failures, monitor failures and their consequences, lessen the likelihood of failures that might cause
harm and take appropriate remedial actions. Buyer will fully indemnify TI and its representatives against any damages arising out of the use
of any TI components in safety-critical applications.
In some cases, TI components may be promoted specifically to facilitate safety-related applications. With such components, TIs goal is to
help enable customers to design and create their own end-product solutions that meet applicable functional safety standards and
requirements. Nonetheless, such components are subject to these terms.
No TI components are authorized for use in FDA Class III (or similar life-critical medical equipment) unless authorized officers of the parties
have executed a special agreement specifically governing such use.
Only those TI components which TI has specifically designated as military grade or enhanced plastic are designed and intended for use in
military/aerospace applications or environments. Buyer acknowledges and agrees that any military or aerospace use of TI components
which have not been so designated is solely at the Buyer's risk, and that Buyer is solely responsible for compliance with all legal and
regulatory requirements in connection with such use.
TI has specifically designated certain components as meeting ISO/TS16949 requirements, mainly for automotive use. In any case of use of
non-designated products, TI will not be responsible for any failure to meet ISO/TS16949.
Audio www.ti.com/audio Automotive and Transportation www.ti.com/automotive
Amplifiers amplifier.ti.com Communications and Telecom www.ti.com/communications
Data Converters dataconverter.ti.com Computers and Peripherals www.ti.com/computers
DLP Products www.dlp.com Consumer Electronics www.ti.com/consumer-apps
DSP dsp.ti.com Energy and Lighting www.ti.com/energy
Clocks and Timers www.ti.com/clocks Industrial www.ti.com/industrial
Interface interface.ti.com Medical www.ti.com/medical
Logic logic.ti.com Security www.ti.com/security
Power Mgmt power.ti.com Space, Avionics and Defense www.ti.com/space-avionics-defense
Microcontrollers microcontroller.ti.com Video and Imaging www.ti.com/video
RFID www.ti-rfid.com
OMAP Applications Processors www.ti.com/omap TI E2E Community e2e.ti.com
Wireless Connectivity www.ti.com/wirelessconnectivity
TPA3122D2
www.ti.com
SLOS527A DECEMBER 2007 REVISED DECEMBER 2007
15-W STEREO CLASS-D AUDIO POWER AMPLIFIER

1FEATURES APPLICATIONS
10-W/ch into an 4- Load From a 17-V Supply Televisions
15-W/ch into an 8- Load From a 28-V Supply
Operates from 10 V to 30 V
DESCRIPTION
Efficient Class-D Operation The TPA3122D2 is a 15-W (per channel) efficient,
Class-D audio power amplifier for driving stereo
Four Selectable, Fixed Gain Settings single ended speakers or mono bridge tied load. The
Internal Oscillator (No External Components TPA3122D2 can drive stereo speakers as low as 4.
Required) The efficiency of the TPA3122D2 eliminates the need
Single Ended Analog Inputs for an external heat sink when playing music.
Thermal and Short-Circuit Protection with The gain of the amplifier is controlled by two gain
Auto Recovery Feature select pins. The gain selections are 20, 26, 32, and
36 dB.
20-pin DIP Package
SIMPLIFIED APPLICATION CIRCUIT
TPA3122D2
1 mF
0.22 mF
Left Channel LIN BSR 22 mH 470 mF
Right Channel RIN ROUT
1 mF 0.68 mF
PGNDR
PGNDL 0.68 mF
1 mF
BYPASS LOUT
22 mH 470 mF
AGND BSL
0.22 mF
PVCCL 10 V to 30 V
10 V to 30 V AVCC
PVCCR
VCLAMP
Shutdown
Control SD 1 mF
Mute Control MUTE

GAIN0
GAIN1
} 4-Step Gain
Control
Please be aware that an important notice concerning availability, standard warranty, and use in critical applications of
Texas Instruments semiconductor products and disclaimers thereto appears at the end of this data sheet.
PRODUCTION DATA information is current as of publication date.

Products conform to specifications per the terms of the Texas Copyright 2007, Texas Instruments Incorporated
Instruments standard warranty. Production processing does not
necessarily include testing of all parameters.
TPA3122D2 www.ti.com
These devices have limited built-in ESD protection. The leads should be shorted together or the device placed in conductive foam
during storage or handling to prevent electrostatic damage to the MOS gates.
N (DIP) PACKAGE
(TOP VIEW)
PVCCL 1 20 PGNDL
SD 2 19 LOUT
MUTE 3 18 BSL
LIN 4 17 AVCC
RIN 5 16 AVCC
BYPASS 6 15 GAIN0
AGND 7 14 GAIN1
AGND 8 13 BSR
VCLAMP 9 12 ROUT
PVCCR 10 11 PGNDR
TERMINAL FUNCTIONS
TERMINAL
20-PIN I/O DESCRIPTION
NAME
(DIP)
Shutdown signal for IC (low = disabled, high = operational). TTL logic levels with compliance to
SD 2 I
AVCC.
RIN 5 I Audio input for right channel.
LIN 4 I Audio input for left channel.
GAIN0 15 I Gain select least significant bit. TTL logic levels with compliance to AVCC.
GAIN1 14 I Gain select most significant bit. TTL logic levels with compliance to AVCC.
Mute signal for quick disable/enable of outputs (high = outputs switch at 50% duty cycle; low =
MUTE 3 I
outputs enabled). TTL logic levels with compliance to AVCC.
BSL 18 I/O Bootstrap I/O for left channel.
PVCCL 1 Power supply for left channel H-bridge, not internally connected to PVCCR or AVCC.
LOUT 19 O Class-D -H-bridge positive output for left channel.
PGNDL 20 Power ground for left channel H-bridge.
VCLAMP 9 Internally generated voltage supply for bootstrap capacitors.
BSR 13 I/O Bootstrap I/O for right channel.
ROUT 12 O Class-D -H-bridge negative output for right channel.
PGNDR 11 Power ground for right channel H-bridge.
PVCCR 10 Power supply for right channel H-bridge, not connected to PVCCL or AVCC.
AGND 8 Analog ground for digital/analog cells in core.
AGND 7 Analog Ground for analog cells in core.
Reference for pre-amplifier inputs. Nominally equal to AVCC/8. Also controls start-up time via
BYPASS 6 O
external capacitor sizing.
AVCC 16, 17 High-voltage analog power supply. Not internally connected to PVCCR or PVCCL
2 Submit Documentation Feedback Copyright 2007, Texas Instruments Incorporated
Product Folder Link(s) :TPA3122D2

www.ti.com
TPA3122D2
ABSOLUTE MAXIMUM RATINGS

over operating free-air temperature range (unless otherwise noted) (1)
VALUE UNIT
VCC Supply voltage, AVCC, PVCC 0.3 to 36 V
VI Logic input voltage SD, MUTE, GAIN0, GAIN1 0.3 to VCC +0.3 0.3 to VCC +0.3 V
VIN Analog input voltage RIN, LIN 0.3 to 7 V
Continuous total power dissipation See Dissipation Rating Table
TA Operating free-air temperature range 40 to 85 C
TJ Operating junction temperature range 40 to 150 C
Tstg Storage temperature range -65 to 150 C
RL Load resistance (Minimum value) 3.2 kV
Human body model (all pins) 2 kV
ESD Electrostatic Discharge
Charged-device model (all pins) 500 V
(1) Stresses beyond those listed under absolute maximum ratings may cause permanent damage to the device. These are stress ratings
only, and functional operations of the device at these or any other conditions beyond those indicated under recommended operating
conditions is not implied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability.
DISSIPATION RATINGS
PACKAGE (1) TA 25C DERATING FACTOR TA = 70C TA = 85C
20-pin DIP 1.87 W 15 mW/C 1.20 W 0.97 W
(1) For the most current package and ordering information, see the Package Option Addendum at the end of this document, or see the TI
Web site at www.ti.com.
RECOMMENDED OPERATING CONDITIONS

MIN MAX UNIT
VCC Supply voltage PVCC, AVCC 10 30 V
VIH High-level input voltage SD, MUTE, GAIN0, GAIN1 2 V
VIL Low-level input voltage SD, MUTE, GAIN0, GAIN1 0.8 V
SD, VI = VCC, VCC = 30 V 125
IIH High-level input current MUTE, VI = VCC, VCC = 30 V 125 A
GAIN0, GAIN1, VI = VCC, VCC = 24 V 125
SD, VI = 0, VCC = 30 V 1
IIL Low-level input current MUTE, VI = 0 V, VCC = 30 V 1 A
GAIN0, GAIN1, VI = 0 V, VCC = 24 V 1
TA Operating free-air temperature 40 85 C
Copyright 2007, Texas Instruments Incorporated Submit Documentation Feedback 3

DC CHARACTERISTICS
TA = 25C, VCC = 24 V, RL = 4 (unless otherwise noted)
PARAMETER TEST CONDITIONS MIN TYP MAX UNIT
Class-D output offset voltage
| VOS | VI = 0 V, AV = 36 dB 7.5 50 mV
(measured differentially)
V(BYPASS) Bypass output voltage No load AVCC/8 V
ICC(q) Quiescent supply current SD = 2 V, MUTE = 0 V, No load 23 37 mA
ICC(q) Quiescent supply current in mute mode MUTE = 2 V, No load 23 mA
ICC(q) Quiescent supply current in shutdown 1
SD = 0.8 V , No load 0.39 mA
mode
rDS(on) Drain-source on-state resistance 200 m
Gain0 = 0.8 V 18 20 22
Gain1 = 0.8 V
Gain0 = 2 V 24 26 28
G Gain dB
Gain0 = 0.8 V 30 32 34
Gain1 = 2 V
Gain0 = 2 V 34 36 38
Mute Attenuation VI = 1Vrms 82
AC CHARACTERISTICS
TA = 25C, VCC = 24V, RL = 4 (unless otherwise noted)
PARAMETER TEST CONDITIONS MIN TYP MAX UNIT
VCC = 12 V, Vripple = 200 mVPP 100 Hz 30 dB
KSVR Supply ripple rejection Gain = 20 dB
1 kHz -48 dB
VCC = 12 V, RL = 4 , f = 1 kHz 4
Output Power at 1% THD+N
VCC = 24 V, RL = 8 , f = 1 kHz 8
PO W
Output Power at 10% VCC = 12 V, RL = 4 , f = 1 kHz 5
THD+N VCC = 24 V, RL = 8 , f = 1 kHz 10
Total harmonic distortion + RL = 4 , f = 1 kHz, PO = 1 W 0.1%
THD+N
noise RL = 8 , f = 1 kHz, PO = 1 W 0.06%
85 V
Vn Output integrated noise floor 20 Hz to 22 kHz, A-weighted filter, Gain = 20 dB
80 dB
Crosstalk PO = 1 W, f = 1kHz; Gain = 20 dB 60 dB
SNR Signal-to-noise ratio Max Output at THD+N < 1%, f = 1 kHz, Gain = 20 dB 99 dB
Thermal trip point 150 C
Thermal hysteresis 30 C
fOSC Oscillator frequency 10 V VCC 230 250 270 kHz
mute delay time from mute input switches high until outputs muted 120 msec
t
unmute delay time from mute input switches low until outputs unmuted 120 msec

www.ti.com
TPA3122D2
FUNCTIONAL BLOCK DIAGRAM
BSL
AVCC AVDD PVCCL
REGULATOR
HS
+ LOUT
- VCLAMP
LS
AVDD AVDD PGNDL
LIN
SC
AVDD/2 DETECT
AGND
CONTROL
SD
BIAS
VCLAMP
THERMAL
MUTE
MUTE CONTROL
OSC/RAMP
BYPASS BYPASS
GAIN1 AV
CONTROL
GAIN0
SC
DETECT
BSR
PVCCR
HS
ROUT
-
VCLAMP
+
LS
PGNDR
AVDD
AVDD
RIN
AVDD/2

TYPICAL CHARACTERISTICS
TOTAL HARMONIC DISTORTION + NOISE TOTAL HARMONIC DISTORTION + NOISE
vs vs
FREQUENCY (SE) FREQUENCY (SE)
10 10
Gain = 20 dB Gain = 20 dB
RL = 4 (SE) RL = 4 (SE)
VCC = 12 V VCC = 18 V
PO = 2 W PO = 5 W
1 1
THD+N %
THD+N %
0.1 0.1
PO = 1 W PO = 0.5 W PO = 1 W
PO = 2.5 W
0.01 0.01
20 100 1k 10k 20k 20 100 1k 10k 20k
f Frequency Hz f Frequency Hz G002

G001
Figure 1. Figure 2.

vs vs
10 10
RL = 4 (SE) RL = 8 (SE)
VCC = 24 V VCC = 24 V PO = 5 W
PO = 5 W
1 1
THD+N %
THD+N %
PO = 2.5 W
0.1 0.1
PO = 2.5 W PO = 1 W PO = 1 W
0.01 0.01
20 100 1k 10k 20k 20 100 1k 10k 20k
f Frequency Hz G003
f Frequency Hz G004
Figure 3. Figure 4.

vs vs
OUTPUT POWER (SE) OUTPUT POWER (SE)
10 10
RL = 4 (SE) RL = 8 (SE)
1 1
THD+N %
THD+N %
VCC = 12 V VCC = 12 V
VCC = 18 V
0.1 0.1
VCC = 24 V VCC = 24 V
VCC = 18 V
0.01 0.01
0.01 0.1 1 10 40 0.01 0.1 1 10 40
PO Output Power W G005

Figure 5. Figure 6.

www.ti.com
TPA3122D2
TYPICAL CHARACTERISTICS (continued)
CROSSTALK CROSSTALK
vs vs
0 0
PO = 0.25 W PO = 0.125 W
20 RL = 4 (SE) 20 RL = 8 (SE)
VCC = 18 V VCC = 18 V
Crosstalk dB
Crosstalk dB
40 40
Left to Right
Left to Right
60 60
80 80
Right to Left Right to Left

100 100
20 100 1k 10k 20k 20 100 1k 10k 20k
f Frequency Hz G007
f Frequency Hz G008
Figure 7. Figure 8.
GAIN/PHASE GAIN/PHASE
vs vs
400 30 200
Gain
20 25 100
Gain
200 20
Gain dBr A
Gain dBr A
0 0
Phase
Phase
Phase Phase
15
100
20
0 10
Gain = 20 dB L filt = 22 mH Gain = 20 dB L filt = 47 mH
PO = 0.125 W Cfilt = 0.68 mF PO = 0.125 W Cfilt = 0.22 mF 200
5
40 RL = 4 (SE) Cdc = 470 mF RL = 8 (SE) Cdc = 470 mF
VCC = 24 V VCC = 18 V
200 0 300
100 1k 10k 100k 20 100 1k 10k 200k
f Frequency Hz G009
f Frequency Hz G010
Figure 9. Figure 10.
OUTPUT POWER OUTPUT POWER

vs vs
SUPPLY VOLTAGE (SE) SUPPLY VOLTAGE (SE)
15 18
16
R L = 4 W (SE) RL = 8 (SE)
PO Output Power W
14
PO Output Power W
10 12
THD+N = 10%
10
THD+N = 10%
8
5 6 THD+N = 1%
THD+N = 1%
4
2
0 0
10 12 14 16 18 20 10 12 14 16 18 20 22 24 26 28 30
PVCC Supply Voltage V G011 PVCC Supply Voltage V G012
Figure 12.
NOTE: Dashed line = Thermally limited
Figure 11.

EFFICIENCY EFFICIENCY
vs vs
OUTPUT POWER (SE) OUTPUT POWER (SE)
100 100
80 80
Efficiency %
Efficiency %
60 60 VCC = 24 V
VCC = 12 V
40 40
VCC = 18 V
20 Gain = 20 dB 20
RL = 4 (SE) Gain = 20 dB
VCC = 12 V RL = 8 (SE)
0 0
0 1 2 3 4 5 6 7 0 2 4 6 8 10 12 14
PO Output Power W PO Output Power W

G013 G014
SUPPLY CURRENT SUPPLY CURRENT

vs vs
TOTAL OUTPUT POWER (SE) TOTAL OUTPUT POWER (SE)
1.4 0.7
1.2 RL = 4 (SE) 0.6 RL = 8 (SE)
VCC = 12 V
ICC Supply Current A
1.0 0.5
0.8 0.4
VCC = 24 V
0.6 0.3
0.4 0.2 VCC = 18 V
0.2 0.1
VCC = 12 V
0.0 0.0
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
PO Total Output Power W G015

PO Total Output Power W G016
POWER SUPPLY REJECTION RATIO TOTAL HARMONIC DISTORTION + NOISE

vs vs
FREQUENCY (SE) FREQUENCY (BTL)
0 10
Gain = 20 dB
20 RL = 8 (BTL)
VCC = 24 V
1 PO = 20 W
40
THD+N %
PSRR dB
PO = 1 W
60 0.1
80
Gain = 20 dB 0.01 PO = 5 W
RL = 4 (SE)
100
VCC = 12 V
Vripple = 200 mVp-p
120 0.001
20 100 1k 10k 20k 20 100 1k 10k 20k
f Frequency Hz G017
f Frequency Hz G018

www.ti.com
TPA3122D2
TOTAL HARMONIC DISTORTION + NOISE GAIN/PHASE

vs vs
OUTPUT POWER (BTL) FREQUENCY (BTL)
10 30
Gain = 20 dB Phase
RL = 8 (BTL) 20
200
1 VCC = 12 V Gain
300
10
Gain dBr A
THD+N %
Phase
400
0.1 0
500
10
0.01 Gain = 20 dB L filt = 33 mH
VCC = 18 V PO = 0.125 W Cfilt = 1 mF 600
VCC = 24 V 20
RL = 8 (BTL)
VCC = 24 V
0.001 30 700
0.01 0.1 1 10 50 20 100 1k 10k 200k
PO Output Power W f Frequency Hz G020

G019
OUTPUT POWER EFFICIENCY

vs vs
SUPPLY VOLTAGE (BTL) OUTPUT POWER (BTL)
70 100
Gain = 20 dB
60 RL = 8 (BTL)
80
PO Output Power W
50
Efficiency %
60 VCC = 24 V
40 THD+N = 10% VCC = 12 V
30 VCC = 18 V
40
THD+N = 1%
20
20
10 Gain = 20 dB
RL = 8 (BTL)
0 0
10 12 14 16 18 20 22 24 26 28 30 0 4 8 12 16 20 24 28
PVCC Supply Voltage V G021

SUPPLY CURRENT POWER SUPPLY REJECTION RATIO

vs vs
TOTAL OUTPUT POWER (BTL) FREQUENCY (BTL)
2.0 0
1.8
RL = 8 (BTL) 20 RL = 8 (BTL)
1.6 VCC = 18 V
VCC = 24 V
1.4 Vripple = 200 mVp-p
40
PSRR dB
1.2 VCC = 12 V
1.0 60
0.8
80
0.6
VCC = 24 V
0.4
100
0.2
0.0 120
0 4 8 12 16 20 24 28 20 100 1k 10k 20k
PO Total Output Power W f Frequency Hz G024

G023

APPLICATION INFORMATION
CLASS-D OPERATION
This section focuses on the class-D operation of the TPA3122D2.
Traditional Class-D Modulation Scheme

The TPA3122D2 operates in AD mode. There are two main configurations that may be used. For stereo
operation, the TPA3122D2 should be configured in a single-ended (SE) half bridge amplifier. For mono
applications, TPA3122D2 may be used as a bridge tied load (BTL) amplifier. The traditional class-D modulation
scheme, which is used in the TPA3122D2 BTL configuration, has a differential output where each output is 180
degrees out of phase and changes from ground to the supply voltage, VCC. Therefore, the differential pre-filtered
output varies between positive and negative VCC, where filtered 50% duty cycle yields
0 V across the load. The traditional class-D modulation scheme with voltage and current waveforms is shown in
Figure 25.
+12 V
OUTP
0V
-12 V
OUTN
0V
+12 V
Differential Voltage
0V
Across Load
-12 V
Current
Figure 25. Traditional Class-D Modulation Scheme's Output Voltage and Current Waveforms into an
Inductive Load With No Input
Supply Pumping
One issue encountered in single-ended (SE) class-D amplifier designs is supply pumping. Power-supply pumping
is a rise in the local supply voltage due to energy being driven back to the supply by operation of the class-D
amplifier. This phenomenon is most evident at low audio frequencies and when both channels are operating at
the same frequency and phase. At low levels, power-supply pumping results in distortion in the audio output due
to fluctuations in supply voltage. At higher levels, pumping can cause the overvoltage protection to operate,
which temporarily shuts down the audio output.
Several things can be done to relieve power-supply pumping. The lowest impact is to operate the two inputs out
of phase 180 and reverse the speaker connections. Because most audio is highly correlated, this causes the
supply pumping to be out of phase and not as severe. If this is not enough, the amount of bulk capacitance on
the supply must be increased. Also, improvement is realized by hooking other supplies to this node, thereby,
sinking some of the excess current. Power-supply pumping should be tested by operating the amplifier at low
frequencies and high output levels.
Gain setting via GAIN0 and GAIN1 inputs

The gain of the TPA3122D2 is set by two input terminals, GAIN0 and GAIN1.
The gains listed in Table 1 are realized by changing the taps on the input resistors and feedback resistors inside
the amplifier. This causes the input impedance (ZI) to be dependent on the gain setting. The actual gain settings
are controlled by ratios of resistors, so the gain variation from part-to-part is small. However, the input impedance
from part-to-part at the same gain may shift by 20% due to shifts in the actual resistance of the input resistors.

www.ti.com
TPA3122D2
For design purposes, the input network (discussed in the next section) should be designed assuming an input
impedance of 8 k, which is the absolute minimum input impedance of the TPA3122D2. At the higher gain
settings, the input impedance could increase as high as 72 k
Table 1. Gain Setting

AMPLIFIER GAIN (dB) INPUT IMPEDANCE (k)
GAIN1 GAIN0
TYPICAL TYPICAL
0 0 20 60
0 1 26 30
1 0 32 15
1 1 36 9
INPUT RESISTANCE
Changing the gain setting can vary the input resistance of the amplifier from its smallest value, 10 k 20%, to
the largest value, 60 k 20%. As a result, if a single capacitor is used in the input high-pass filter, the -3 dB or
cutoff frequency may change when changing gain steps.
Zf
Ci
Zi
Input IN
Signal
The -3-dB frequency can be calculated using Equation 1. Use the ZI values given in Table 1.
1
f =
2p Zi Ci (1)
INPUT CAPACITOR, CI
In the typical application, an input capacitor I) is required to allow the amplifier to bias the input signal to the
proper dc level for optimum operation. In this case, CI and the input impedance of the amplifier (ZI) form a
high-pass filter with the corner frequency determined in Equation 2.
3 dB
1
fc =
2p Zi Ci
fc (2)
The value of CI is important, as it directly affects the bass (low-frequency) performance of the circuit. Consider
the example where ZI is 20 k and the specification calls for a flat bass response down to 20 Hz. Equation 2 is
reconfigured as Equation 3.
1
Ci =
2p Zi fc (3)
In this example, CI is 0.4 F; so, one would likely choose a value of 0.47 F as this value is commonly used. If
the gain is known and is constant, use ZI from Table 1 to calculate CI. A further consideration for this capacitor is
the leakage path from the input source through the input network I) and the feedback network to the load. This
leakage current creates a dc offset voltage at the input to the amplifier that reduces useful headroom, especially

in high gain applications. For this reason, a low-leakage tantalum or ceramic capacitor is the best choice. When
polarized capacitors are used, the positive side of the capacitor should face the amplifier input in most
applications as the dc level there is held at 2 V, which is likely higher than the source dc level. Note that it is
important to confirm the capacitor polarity in the application. Additionally, lead-free solder can create dc offset
voltages and it is important to ensure that boards are cleaned properly.
Single Ended Output Capacitor, Co

In single ended (SE) applications, the DC blocking capacitor forms a high pass filter with speaker impedance.
The frequency response rolls of with decreasing frequency at a rate of 20dB/decade. The cutoff frequency is
determined by
fc = 1/2CoZL
Table 2 shows some common component values and the associated cutoff frequencies:
Table 2. Common Filter Responses

CSE DC Blocking Capacitor (F)
Speaker Impedance ()
fc = 60 Hz fc = 40 Hz fc = 20 Hz
4 680 1000 2200
8 330 470 1000
Output Filter and Frequency Response

For the best frequency response, a flat passband output filter (second order Butterworth) may be used. The
output filter components consist of the series inductor and capacitor to ground at the LOUT and ROUT pins.
There are several possible configurations depending on the speaker impedance and whether the output
configuration is Single Ended (SE) or Bridge Tied Load (BTL). Table 3 list several possible arrangements.
Table 3. Recommended Filter Output Components

Output Configuration Speaker Impedance () Filter Inductor (H) Filter Capacitor (nF)
4 22 680
Single Ended (SE)
8 47 390
4 10 1500
Bridge Tied Load
8 22 680
Lfilter LOUT / ROUT Lfilter

LOUT
Cfilter Cfilter
ROUT Lfilter
Cfilter
Figure 26. BTL Filter Configuration Figure 27. SE Filter Configuration

www.ti.com
TPA3122D2
Power Supply Decoupling, CS

The TPA3122D2 is a high-performance CMOS audio amplifier that requires adequate power supply decoupling
to ensure that the output total harmonic distortion (THD) is as low as possible. Power supply decoupling also
prevents oscillations for long lead lengths between the amplifier and the speaker. The optimum decoupling is
achieved by using two capacitors of different types that target different types of noise on the power supply leads.
For higher frequency transients, spikes, or digital hash on the line, a good low equivalent-series-resistance (ESR)
ceramic capacitor, typically 0.1 F to 1 F placed as close as possible to the device VCC lead works best. For
filtering lower frequency noise signals, a larger aluminum electrolytic capacitor of 220 F or greater placed near
the audio power amplifier is recommended. The 220-F capacitor also serves as local storage capacitor for
supplying current during large signal transients on the amplifier outputs. The PVCC terminals provide the power
to the output transistors, so a 220-F or larger capacitor should be placed on each PVCC terminal. A 10-F
capacitor on the AVCC terminal is adequate.
BSN and BSP Capacitors

The half H-bridge output stages use only NMOS transistors. Therefore, they require bootstrap capacitors for the
high side of each output to turn on correctly. A 220-nF ceramic capacitor, rated for at least 25 V, must be
connected from each output to its corresponding bootstrap input. Specifically, one 220-nF capacitor must be
connected from LOUT to BSL, and one 220-nF capacitor must be connected from ROUT to BSR.
The bootstrap capacitors connected between the BSx pins and corresponding output function as a floating power
supply for the high-side N-channel power MOSFET gate drive circuitry. During each high-side switching cycle,
the bootstrap capacitors hold the gate-to-source voltage high enough to keep the high-side MOSFETs turned on.
VCLAMP Capacitor
To ensure that the maximum gate-to-source voltage for the NMOS output transistors is not exceeded, one
internal regulator clamps the gate voltage. One 1-F capacitor must be connected from VCLAMP (pin 11 for
PWP and pin 9 for DIP package) to ground and must be rated for at least 16 V. The voltages at the VCLAMP
terminal may vary with VCC and may not be used for powering any other circuitry.
VBYP Capacitor Selection

The scaled supply reference (VBYP) nominally provides an AVcc/8 internal bias for the preamplifier stages. The
external capacitor for this reference CBYP) is a critical component and serves several important functions. During
start-up or recovery from shutdown mode, CBSP determines the rate at which the amplifier starts up. The second
function is to reduce noise produced by the power supply caused by coupling with the output drive signal. This
noise could result in degraded PSRR and THD + N.
The circuit is designed for a CBSP value of 1 F for best pop performance. The inputs caps should be the same
value. A ceramic or tantalum low-ESR capacitor is recommended.
SHUTDOWN OPERATION
The TPA3122D2 employs a shutdown mode of operation designed to reduce supply current (ICC) to the absolute
minimum level during periods of non-use for power conservation. The SHUTDOWN input terminal should be held
high (see specification table for trip point) during normal operation when the amplifier is in use. Pulling
SHUTDOWN low causes the outputs to mute and the amplifier to enter a low-current state. Never leave
SHUTDOWN unconnected, because amplifier operation would be unpredictable.
For the best power-up pop performance, place the amplifier in the shutdown or mute mode prior to applying the
power supply voltage.
MUTE Operation
The MUTE pin is an input for controlling the output state of the TPA3122D2. A logic high on this terminal causes
the outputs to run at a constant 50% duty cycle. A logic low on this pin enables the outputs. This terminal may be
used as a quick disable/enable of outputs when changing channels on a television or switching between different
audio sources.
The MUTE terminal should never be left floating. For power conservation, the SHUTDOWN terminal should be
used to reduce the quiescent current to the absolute minimum level.

USING LOW-ESR CAPACITORS

Low-ESR capacitors are recommended throughout this application section. A real (as opposed to ideal) capacitor
can be modeled simply as a resistor in series with an ideal capacitor. The voltage drop across this resistor
minimizes the beneficial effects of the capacitor in the circuit. The lower the equivalent value of this resistance,
the more the real capacitor behaves like an ideal capacitor.
SHORT-CIRCUIT PROTECTION
The TPA3122D2 has short-circuit protection circuitry on the outputs that prevents damage to the device during
output-to-output shorts and output-to-GND shorts. When a short circuit is detected on the outputs, the part
immediately disables the output drive. This is an unlatched fault. Normal operation is restored when the fault is
removed.
THERMAL PROTECTION
Thermal protection on the TPA3122D2 prevents damage to the device when the internal die temperature
exceeds 150C. There is a 15C tolerance on this trip point from device to device. Once the die temperature
exceeds the thermal set point, the device enters into the shutdown state and the outputs are disabled. This is not
a latched fault. The thermal fault is cleared once the temperature of the die is reduced by 30C. The device
begins normal operation at this point with no external system interaction.
PRINTED-CIRCUIT BOARD (PCB) LAYOUT

Because the TPA3122D2 is a class-D amplifier that switches at a high frequency, the layout of the printed-circuit
board (PCB) should be optimized according to the following guidelines for the best possible performance.
Decoupling capacitorsThe high-frequency 0.1F decoupling capacitors should be placed as close to the
PVCC (pins 1 and 10) and AVCC (pins 16 and 17) terminals as possible. The VBYP (pin 6) capacitor and
VCLAMP (pin 9) capacitor should also be placed as close to the device as possible. Large (220 F or
greater) bulk power supply decoupling capacitors should be placed near the TPA3122D2 on the PVCCL and
PVCCR terminals.
GroundingThe AVCC (pins 16 and 17) decoupling capacitor and VBYP (pin 6) capacitor should each be
grounded to analog ground (AGND, pins 7 and 8). The PVCCx decoupling capacitors and VCLAMP
capacitors should each be grounded to power ground (PGND, pins 11 and 20). Analog ground and power
ground should be connected at the thermal pad, which should be used as a central ground connection or star
ground for the TPA3122D2.
Output filterThe EMI filter (L1, L2, C9, and C16) should be placed as close to the output terminals as
possible for the best EMI performance. The capacitors should be grounded to power ground.
For an example layout, see the TPA3122D2 Evaluation Module (TPA3122D2EVM) User Manual, (SLOU214).
Both the EVM user manual and the thermal pad application note are available on the TI Web site at
http://www.ti.com.
VCC
22uH
Shutdown Control
0.1uF
Mute Control
470uF
1 PVCCL PGNDL 20
1.0uF 2 19 LEFT_OUT 4.7K 0.68uF
Left Input SD LOUT
3 MUTE BSL 18
4 17 0.22uF
LIN AVCC1
5 RIN AVCC2 16
Right Input 6 15
BYPASS GAIN0
7 AGND1 GAIN1 14
1.0uF 8 13 0.22uF
1.0uF AGND2 BSR RIGHT_OUT
9 VCLAMP ROUT 12
10 PVCCR PGNDR 11
TPA3122_PDIP
4.7K 0.68uF
0.1uF 22uH
470uF
1.0uF 470uF 470uF

0.1uF 10uF
Figure 28. SE 4- Application Schematic

www.ti.com
TPA3122D2
VCC
22uH
Shutdown Control
0.1uF
Mute Control
1 PVCCL PGNDL 20
1.0uF 2 19 LEFT_OUT 4.7K 0.68uF
Plus Input SD LOUT
3 MUTE BSL 18
4 17 0.22uF
LIN AVCC1
5 RIN AVCC2 16
Minus Input 6 15
BYPASS GAIN0
7 AGND1 GAIN1 14
1.0uF 8 13 0.22uF
1.0uF AGND2 BSR RIGHT_OUT
9 VCLAMP ROUT 12
10 PVCCR PGNDR 11
TPA3122_PDIP
4.7K 0.68uF
0.1uF 22uH
1.0uF 470uF 470uF

0.1uF 10uF
Figure 29. BTL 8- Application Schematic
BASIC MEASUREMENT SYSTEM

This application note focuses on methods that use the basic equipment listed below:
Audio analyzer or spectrum analyzer
Digital multimeter (DMM)
Oscilloscope
Twisted-pair wires
Signal generator
Power resistor(s)
Linear regulated power supply
Filter components
EVM or other complete audio circuit
Figure 30 shows the block diagrams of basic measurement systems for class-AB and class-D amplifiers. A sine
wave is normally used as the input signal because it consists of the fundamental frequency only (no other
harmonics are present). An analyzer is then connected to the APA output to measure the voltage output. The
analyzer must be capable of measuring the entire audio bandwidth. A regulated dc power supply is used to
reduce the noise and distortion injected into the APA through the power pins. A System Two audio measurement
system (AP-II) (Reference 1) by Audio Precision includes the signal generator and analyzer in one package.
The generator output and amplifier input must be ac-coupled. However, the EVMs already have the ac-coupling
capacitors, CIN), so no additional coupling is required. The generator output impedance should be low to avoid
attenuating the test signal, and is important because the input resistance of APAs is not high. Conversely, the
analyzer-input impedance should be high. The output resistance, ROUT, of the APA is normally in the hundreds of
milliohms and can be ignored for all but the power-related calculations.
Figure 30(a) shows a class-AB amplifier system. It takes an analog signal input and produces an analog signal
output. This amplifier circuit can be directly connected to the AP-II or other analyzer input.
This is not true of the class-D amplifier system shown in Figure 30(b), which requires low-pass filters in most
cases in order to measure the audio output waveforms. This is because it takes an analog input signal and
converts it into a pulse-width modulated (PWM) output signal that is not accurately processed by some
analyzers.

Power Supply
Signal APA RL Analyzer

Generator 20 Hz - 20 kHz
(a) Basic Class-AB
Power Supply
Lfilt
Signal Class-D APA Cfilt Analyzer

Generator RL 20 Hz - 20 kHz
(b) Traditional Class-D
Figure 30. Audio Measurement Systems

www.ti.com
TPA3122D2
SE Input and SE Output (TPA3122D2 Stereo Configuration)

The SE input and output configuration is used with class-AB amplifiers. A block diagram of a fully SE
measurement circuit is shown in Figure 31. SE inputs normally have one input pin per channel. In some cases,
two pins are present; one is the signal and the other is ground. SE outputs have one pin driving a load through
an output ac coupling capacitor and the other end of the load is tied to ground. SE inputs and outputs are
considered to be unbalanced, meaning one end is tied to ground and the other to an amplifier input/output.
The generator should have unbalanced outputs, and the signal should be referenced to the generator ground for
best results. Unbalanced or balanced outputs can be used when floating, but they may create a ground loop that
will effect the measurement accuracy. The analyzer should have balanced inputs to cancel out any
common-mode noise in the measurement.
Evaluation Module
Audio Power
Generator Analyzer
Amplifier
CIN
Lfilt
RGEN RIN CL
VGEN
RANA CANA
Cfilt RL
RANA CANA
Figure 31. SE InputSE Output Measurement Circuit
The following general rules should be followed when connecting to APAs with SE inputs and outputs:
Use an unbalanced source to supply the input signal.
Use twisted pair wire for all connections.
Ensure the cables from the power supply to the APA, and from the APA to the load, can handle the large
currents (see Table 4)

DIFFERENTIAL INPUT AND BTL OUTPUT (TPA3122D2 Mono Configuration)

Many of the class-D APAs and many class-AB APAs have differential inputs and bridge-tied load (BTL) outputs.
Differential inputs have two input pins per channel and amplify the difference in voltage between the pins.
Differential inputs reduce the common-mode noise and distortion of the input circuit. BTL is a term commonly
used in audio to describe differential outputs. BTL outputs have two output pins providing voltages that are 180
degrees out of phase. The load is connected between these pins. This has the added benefits of quadrupling the
output power to the load and eliminating a dc blocking capacitor.
A block diagram of the measurement circuit is shown in Figure 32. The differential input is a balanced input,
meaning the positive (+) and negative (-) pins have the same impedance to ground. Similarly, the SE output
equates to a balanced output.
Evaluation Module
Audio Power
Generator Analyzer
Amplifier
CIN Lfilt
RGEN RIN Cfilt RANA CANA

VGEN RL
CIN Lfilt
RGEN RIN Cfilt RANA CANA
Figure 32. Differential Input, BTL Output Measurement Circuit
The generator should have balanced outputs, and the signal should be balanced for best results. An unbalanced
output can be used, but it may create a ground loop that affects the measurement accuracy. The analyzer must
also have balanced inputs for the system to be fully balanced, thereby cancelling out any common-mode noise in
the circuit and providing the most accurate measurement.
The following general rules should be followed when connecting to APAs with differential inputs and BTL outputs:
Use a balanced source to supply the input signal.
Use twisted-pair wire for all connections.
Ensure that the cables from the power supply to the APA, and from the APA to the load, can handle the large
currents (see Table 4).
Table 4 shows the recommended wire size for the power supply and load cables of the APA system. The real
concern is the dc or ac power loss that occurs as the current flows through the cable. These recommendations
are based on 12-inch long wire with a 20-kHz sine-wave signal at 25C.
Table 4. Recommended Minimum Wire Size for Power Cables

DC POWER LOSS AC POWER LOSS
POUT (W) RL() AWG Size
(MW) (MW)
10 4 18 22 16 40 18 42
2 4 18 22 3.2 8 3.7 8.5
1 8 22 28 2 8 2.1 8.1
< 0.75 8 22 28 1.5 6.1 1.6 6.2

PACKAGE OPTION ADDENDUM
www.ti.com 10-Apr-2008
PACKAGING INFORMATION
Orderable Device Status (1) Package Package Pins Package Eco Plan (2) Lead/Ball Finish MSL Peak Temp (3)
Type Drawing Qty
TPA3122D2N ACTIVE PDIP N 20 20 Pb-Free CU NIPDAU N / A for Pkg Type
(RoHS)
(1)
The marketing status values are defined as follows:
ACTIVE: Product device recommended for new designs.
LIFEBUY: TI has announced that the device will be discontinued, and a lifetime-buy period is in effect.
NRND: Not recommended for new designs. Device is in production to support existing customers, but TI does not recommend using this part in
a new design.
PREVIEW: Device has been announced but is not in production. Samples may or may not be available.
OBSOLETE: TI has discontinued the production of the device.
(2)
Eco Plan - The planned eco-friendly classification: Pb-Free (RoHS), Pb-Free (RoHS Exempt), or Green (RoHS & no Sb/Br) - please check
http://www.ti.com/productcontent for the latest availability information and additional product content details.
TBD: The Pb-Free/Green conversion plan has not been defined.
Pb-Free (RoHS): TI's terms "Lead-Free" or "Pb-Free" mean semiconductor products that are compatible with the current RoHS requirements
for all 6 substances, including the requirement that lead not exceed 0.1% by weight in homogeneous materials. Where designed to be soldered
at high temperatures, TI Pb-Free products are suitable for use in specified lead-free processes.
Pb-Free (RoHS Exempt): This component has a RoHS exemption for either 1) lead-based flip-chip solder bumps used between the die and
package, or 2) lead-based die adhesive used between the die and leadframe. The component is otherwise considered Pb-Free (RoHS
compatible) as defined above.
Green (RoHS & no Sb/Br): TI defines "Green" to mean Pb-Free (RoHS compatible), and free of Bromine (Br) and Antimony (Sb) based flame
retardants (Br or Sb do not exceed 0.1% by weight in homogeneous material)
(3)
MSL, Peak Temp. -- The Moisture Sensitivity Level rating according to the JEDEC industry standard classifications, and peak solder
temperature.
Important Information and Disclaimer:The information provided on this page represents TI's knowledge and belief as of the date that it is
provided. TI bases its knowledge and belief on information provided by third parties, and makes no representation or warranty as to the
accuracy of such information. Efforts are underway to better integrate information from third parties. TI has taken and continues to take
reasonable steps to provide representative and accurate information but may not have conducted destructive testing or chemical analysis on
incoming materials and chemicals. TI and TI suppliers consider certain information to be proprietary, and thus CAS numbers and other limited
information may not be available for release.
In no event shall TI's liability arising out of such information exceed the total purchase price of the TI part(s) at issue in this document sold by TI
to Customer on an annual basis.
Addendum-Page 1
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications, enhancements, improvements,
and other changes to its products and services at any time and to discontinue any product or service without notice. Customers should
obtain the latest relevant information before placing orders and should verify that such information is current and complete. All products are
sold subject to TIs terms and conditions of sale supplied at the time of order acknowledgment.
TI warrants performance of its hardware products to the specifications applicable at the time of sale in accordance with TIs standard
warranty. Testing and other quality control techniques are used to the extent TI deems necessary to support this warranty. Except where
mandated by government requirements, testing of all parameters of each product is not necessarily performed.
TI assumes no liability for applications assistance or customer product design. Customers are responsible for their products and
applications using TI components. To minimize the risks associated with customer products and applications, customers should provide
TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right, copyright, mask work right,
or other TI intellectual property right relating to any combination, machine, or process in which TI products or services are used. Information
published by TI regarding third-party products or services does not constitute a license from TI to use such products or services or a
warranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual
property of the third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of TI information in TI data books or data sheets is permissible only if reproduction is without alteration and is accompanied
by all associated warranties, conditions, limitations, and notices. Reproduction of this information with alteration is an unfair and deceptive
business practice. TI is not responsible or liable for such altered documentation. Information of third parties may be subject to additional
restrictions.
Resale of TI products or services with statements different from or beyond the parameters stated by TI for that product or service voids all
express and any implied warranties for the associated TI product or service and is an unfair and deceptive business practice. TI is not
responsible or liable for any such statements.
TI products are not authorized for use in safety-critical applications (such as life support) where a failure of the TI product would reasonably
be expected to cause severe personal injury or death, unless officers of the parties have executed an agreement specifically governing
such use. Buyers represent that they have all necessary expertise in the safety and regulatory ramifications of their applications, and
acknowledge and agree that they are solely responsible for all legal, regulatory and safety-related requirements concerning their products
and any use of TI products in such safety-critical applications, notwithstanding any applications-related information or support that may be
provided by TI. Further, Buyers must fully indemnify TI and its representatives against any damages arising out of the use of TI products in
such safety-critical applications.
TI products are neither designed nor intended for use in military/aerospace applications or environments unless the TI products are
specifically designated by TI as military-grade or "enhanced plastic." Only products designated by TI as military-grade meet military
specifications. Buyers acknowledge and agree that any such use of TI products which TI has not designated as military-grade is solely at
the Buyer's risk, and that they are solely responsible for compliance with all legal and regulatory requirements in connection with such use.
TI products are neither designed nor intended for use in automotive applications or environments unless the specific TI products are
designated by TI as compliant with ISO/TS 16949 requirements. Buyers acknowledge and agree that, if they use any non-designated
products in automotive applications, TI will not be responsible for any failure to meet such requirements.
Following are URLs where you can obtain information on other Texas Instruments products and application solutions:
Amplifiers amplifier.ti.com Audio www.ti.com/audio
Data Converters dataconverter.ti.com Automotive www.ti.com/automotive
DLP Products www.dlp.com Broadband www.ti.com/broadband
DSP dsp.ti.com Digital Control www.ti.com/digitalcontrol
Clocks and Timers www.ti.com/clocks Medical www.ti.com/medical
Interface interface.ti.com Military www.ti.com/military
Logic logic.ti.com Optical Networking www.ti.com/opticalnetwork
Power Mgmt power.ti.com Security www.ti.com/security
Microcontrollers microcontroller.ti.com Telephony www.ti.com/telephony
RFID www.ti-rfid.com Video & Imaging www.ti.com/video
RF/IF and ZigBee Solutions www.ti.com/lprf Wireless www.ti.com/wireless

TI Class D Amplifier

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

TI Class D Amplifier

Hochgeladen von

Copyright:

Verfügbare Formate

Application Report

SLOA068 October 2001

Guidelines for Measuring Audio Power Amplifier

THD+N versus output power Crosstalk versus frequency

2 Guidelines for Measuring Audio Power Amplifier Performance

2 Basic Measurement System

Audio analyzer or spectrum analyzer Digital multimeter (DMM)

Guidelines for Measuring Audio Power Amplifier Performance 3

Signal APA RL Analyzer

(a) Basic Class-AB Audio Measurement System

Used With Traditional Class-D APAs Only

(b) Filter-Free and Traditional Class-D Audio Measurement System

4 Guidelines for Measuring Audio Power Amplifier Performance

3.1 Differential Input and BTL Output (TPA731 and TPA2000D1)

Twisted-Pair Wire Twisted-Pair Wire

Figure 2. Differential InputBTL Output Measurement Circuit

Use a balanced source to supply the input signal.

Use an analyzer with balanced inputs.

Use twisted-pair wire for all connections.

Guidelines for Measuring Audio Power Amplifier Performance 5

Table 1. Recommended Minimum Wire Size for Power Cables

PO UT(W) RL () AWG DC Power AC Power

3.2 SE Input and SE Output (TPA0211 and TPA711)

6 Guidelines for Measuring Audio Power Amplifier Performance

Twisted-Pair Wire Twisted-Pair Wire

Figure 3. SE InputSE Output Measurement Circuit

Use an unbalanced source to supply the input signal.

Use an analyzer with balanced inputs.

Use twisted pair wire for all connections.

Use shielding when the system environment is noisy.

3.3 Other Configurations

3.4 Class-D RC Low-Pass Filter

Guidelines for Measuring Audio Power Amplifier Performance 7

Load RC Low-Pass Filters AP Analyzer Input

CFILT CANA RANA

CFILT CANA RANA

Figure 4. Measurement Low-Pass Filter Derivation CircuitClass-D APAs

8 Guidelines for Measuring Audio Power Amplifier Performance

Table 2. Typical RC Measurement Filter Values

4 Total Harmonic Distortion Plus Noise (THD+N)

Guidelines for Measuring Audio Power Amplifier Performance 9

AP Generator Out Audio Power Amplifier AP Analyzer In

CFILT Inputs Balanced

Figure 5. THD+N Measurement Circuit Using the AP-II Measurement System:

4.1 THD+N vs Output Power

Figure 6. THD+N vs POUT for the TPA2001D1 and the TPA731

10 Guidelines for Measuring Audio Power Amplifier Performance

Figure 7. THD+N vs Frequency for the TPA2001D1 and the TPA731

4.3 Maximum Output Power Bandwidth

4.4 Maximum Input Voltage

Guidelines for Measuring Audio Power Amplifier Performance 11

5.1 Integrated Noise vs Frequency

Audio Power Amplifier AP Analyzer In

CFILT Inputs Balanced

Outputs Off (No Connect) Regulated

Figure 8. Noise Measurement Circuit

12 Guidelines for Measuring Audio Power Amplifier Performance

Figure 9. Measured Results of Noise Circuit

5.2 Signal-to-Noise Ratio

6 Gain and Phase

Guidelines for Measuring Audio Power Amplifier Performance 13

AP Generator Out Audio Power Amplifier AP Analyzer In

Figure 10. Gain and Phase Measurement Circuit

Figure 11. TPA731 Gain and Phase Measurements