TPA324x and TPA325x Post-Filter Feedback

Application Report
SLAA788A – September 2017 – Revised March 2018
TPA324x and TPA325x Post-Filter Feedback
Dan Kisling, Matthew Beardsworth
ABSTRACT
The TPA324x and TPA325x (TPA3244, TPA3245, TPA3250, TPA3251, TPA3255) Class-D audio amplifier
families deliver high audio performance with less than 0.01% total harmonic distortion and noise (THD+N)
to clipping. The high level of audio performance makes this device an ideal candidate for high resolution
and high fidelity audio applications, which previously could only be achieved by Class-AB amplifiers. The
TPA324x and TPA325x devices are analog input, closed loop (internal feedback network) Class-D audio
amplifiers that can be further enhanced by adding an additional post-filter feedback (PFFB) or PFFB loop.
This application report shows one optional implementation of PFFB for the TPA3244, TPA3245, TPA3250,
TPA3251, and TPA3255 amplifiers.
PFFB offers many benefits including lower output noise, improved THD+N performance, improved IMD
performance, lower output impedance, frequency response less affected by load impedance, and
suppression of nonlinearities of the LC filter.
Contents
1 PFFB Implementation ....................................................................................................... 2
2 Closed Loop Gain............................................................................................................ 4
3 LC Filter Distortion ........................................................................................................... 5
4 Performance Results ........................................................................................................ 6
5 Output Noise ................................................................................................................. 6
6 SNR and DNR ................................................................................................................ 6
7 THD+N vs Power ............................................................................................................ 7
8 THD+N vs Frequency ....................................................................................................... 9
9 SMPTE IMD ................................................................................................................. 11
10 CCIF IMD .................................................................................................................... 12
11 Stability Analysis ........................................................................................................... 13
12 Stability Testing ............................................................................................................. 14
13 Overshoot for Square Wave Input ....................................................................................... 14
14 Calculating Phase Margin ................................................................................................. 15
15 TPA3245 PFFB Stability................................................................................................... 16
16 Frequency Reponse at Full Scale ........................................................................................ 18
Appendix A TPA3244............................................................................................................ 19
Appendix B TPA3245............................................................................................................ 25
Appendix C TPA3250............................................................................................................ 31
Appendix D TPA3251............................................................................................................ 37
Appendix E TPA3255............................................................................................................ 43
Trademarks
All trademarks are the property of their respective owners.
SLAA788A – September 2017 – Revised March 2018 TPA324x and TPA325x Post-Filter Feedback 1
Submit Documentation Feedback
Copyright © 2017–2018, Texas Instruments Incorporated
PFFB Implementation www.ti.com
1 PFFB Implementation
Post filter feedback is implemented by adding a secondary feedback loop external to the amplifier. This
feedback loop takes a fraction of the output voltage signal of the amplifier after the external LC filter and
sends an error signal back to the input of the amplifier. See Figure 1.
GND R_fb_gnd
C_fb_in C_fb_out
R_fb
R_in L_out
In A Out A
TPA324x/5x C_out
Class-D Amplifier
GND
Copyright © 2017, Texas Instruments Incorporated
Figure 1. Post-Filter Feedback Loop
The following components are required for PFFB implementation:

• R_fb – Feedback resistor
• C_fb_in – Capacitor on input side of the feedback network
• C_fb_out – Capacitor on output side of the feedback network
• R_fb_gnd – Resistor between C_fb_in and C_fb_out to GND in the feedback network
• R_in – Input summing junction resistor
• C_z – Zobel network capacitor
• R_z – Zobel network resistor
• C_op – Op-Amp feedback capacitor
• R_op-fb – Op-Amp feedback resistor
2 TPA324x and TPA325x Post-Filter Feedback SLAA788A – September 2017 – Revised March 2018
www.ti.com PFFB Implementation
Figure 2 shows the passive PFFB implementation used throughout the document for TPA3244, TPA3245,
TPA3250, TPA3251, and TPA3255 amplifiers.
GND R_fb_gnd
C_fb_in C_fb_out
R_fb
GND R_fb_gnd
C_fb_in C_fb_out
R_fb
C_op
R_op-fb C_z R_z

10 k
± R_in L_out GND
In A Out A
+ C_out
GND
C_op
R_op-fb C_out
GND
10 k
± R_in L_out
In B Out B
+ C_z R_z
GND
C_op
TPA324x/5x
R_op-fb Class-D Amplifier C_z R_z GND
10 k
± R_in L_out GND
In C Out C
+ C_out
GND
C_op
R_op-fb
C_out
10 k GND
± R_in L_out
In D Out D
+
GND C_z R_z
R_fb
GND
C_fb_in C_fb_out
R_fb_gnd
GND
R_fb
C_fb_in C_fb_out
R_fb_gnd
GND
Figure 2. Passive PFFB Implementation
Closed Loop Gain www.ti.com
Table 1 lists the component values that have improved audio performance and sufficient stability.
Table 1. Recommended PFFB Component Values

TPA3244 TPA3245 TPA3250 TPA3251 TPA3255
EVM Schematic
PFFB (PVDD = 30 V, (PVDD = 30 V, (PVDD = 36 V, (PVDD = 36 V, (PVDD = 51 V,
Location and
Designator Fpwm = 450 Fpwm = 600 Fpwm = 450 Fpwm = 600 Fpwm = 450
Designator
kHz) kHz) kHz) kHz) kHz)
L_out L2, L3, L4, L5 10 µH 10 µH 10 µH 7 µH 10 µH
C24, C35, C43,
C_out 1 µF 1 µF 1 µF 680 nF 1 µF
C59
R47, R49, R50,
R_fb 18 kΩ 18 kΩ 18 kΩ 18 kΩ 33 kΩ
R51
C_fb_in N/A 220 pF 220 pF 220 pF 220 pF 220 pF
C_fb_out N/A 220 pF 1 nF 220 pF 1 nF 220 pF
R_fb_gnd N/A 2 kΩ 2 kΩ 2 kΩ 2 kΩ 10 kΩ
R4, R12, R44,
R_in 2.7 kΩ 2.7 kΩ 2.7 kΩ 2.7 kΩ 2.7 kΩ
R46
C77, C78, C79,
C_z 220 nF 220 nF 220 nF 220 nF 220 nF
C80
R54, R55, R56,
R_z 1Ω 1Ω 1Ω 1Ω 1Ω
R57
C18, C23, C57,
C_op 330 pF 330 pF 330 pF 330 pF 330 pF
C65
R8, R41, R21,
R_op_fb 10 kΩ 10 kΩ 10 kΩ 10 kΩ 10 kΩ
R25
• R_fb and R_in controls the amount of negative feedback used in this system.
• C_fb_in, C_fb_out, and R_fb_gnd make a feedback network that helps with open load response and
overall network stability.
• C_z and R_z create the Zobel Network and are required for network stability.
The Zobel network helps attenuate the high-frequency ringing by damping the amplitude response of the
output of the LC filter and lowering the quality factor (Q). This is especially required for open-load stability
where the Q of the LC filter is extremely large due to the lack of dampening from the load impedance.
Reducing the Q of the output filter also increases stability, by relaxing the phase shift associated with a
high Q system.
It is best to keep the feedback components close to the device so the feedback signals are not subject to
significant distortion or noise.
2 Closed Loop Gain

With the components for PFFB selected and the gain on the amplifier known, the new closed loop gain
can be estimated. Equation 1 shows the closed-loop gain.
A0
Af
1 A 0E (1)
Use Equation 2 to calculate the feedback factor.
R _ in
E
R _ in R _ fb (2)
www.ti.com LC Filter Distortion
Using Equation 1 and Equation 2, the PFFB gain and the negative feedback gain can be calculated.
Equation 3 shows an example for the TPA3251.
2.7 k
A 0 20 dB 10 E 0.13
2.7 k 18 k
10
Af 4.35 12.8 dB
1 10 u 0.13
where
• The closed loop gain has been reduced to 12.8 dB due to PFFB
• 7.2 dB of PFFB has been applied to the amplifier (3)
Approximately 6 to 7 dB of PFFB has been applied to each amplifier. Table 2 lists the results.
Table 2. PFFB Parameters

Feedback
(PVDD = 30 V, (PVDD = 30 V, (PVDD = 36 V, (PVDD = 36 V, (PVDD = 51 V,
Parameters
Fpwm = 450 kHz) Fpwm = 600 kHz) Fpwm = 450 kHz) Fpwm = 600 kHz) Fpwm = 450 kHz)
Gain (dB) 18 18 20 20 21.5
Feedback factor 0.130 0.130 0.130 0.130 0.076
PFFB gain Af (dB) 11.82 11.82 12.75 12.75 15.93
Negative feedback 6.18 6.18 7.25 7.25 5.57
3 LC Filter Distortion
The LC filter extracts a continuous analog audio signal from the PWM output for the Class-D to suppress
radiating EMI and ripple current in the load connected. However, since the TPA324x and TPA325x
families offer such a high level of performance, the inductor used in the LC filter is the primary contributor
to distortion. By feeding back the output after the inductor in PFFB, inductor and capacitor distortion can
be reduced significantly. For systems where lower distortion is not a requirement, PFFB can allow for the
use of a smaller and less expensive inductor or capacitor and correct for some added distortion. Since
smaller inductors are usually less linear and cause higher distortion, the distortion improvement offered by
PFFB can allow very good system performance even with an inductor of this type. The same can be
stated for smaller and cheaper capacitors.
Figure 3 shows the performance difference between two different 10-µH inductors on the TPA3245. With
PFFB, not only does the overall performance of the amplifier improve, but the performance gap between
the two inductors has been reduced.
10
TPA3245, 4 :, BTL, 1 kHz, MA5172-AE, PFFB
TPA3245, 4 :, BTL, 1 kHz, MA5172-AE, No PFFB
TPA3245, 4 :, BTL, 1 kHz, 7443551730, PFB
TPA3245, 4 :, BTL, 1 kHz, 7443551730, No PFB
1
THD+N (%)
0.1
0.01
0.001
1m 10m 100m 1 10 100
Output Power (W) D001
Figure 3. THD+N vs Power Inductor Comparison
Performance Results www.ti.com
4 Performance Results
PFFB is implemented on the TPA3245, TPA3251, TPA3255, TPA3244, and TPA3250 EVM. The
performance of the EVM is measured, and then the performance of the EVM with PFFB is implemented
for the same conditions. The following measurements were completed for all devices. For select audio
performance results, only the TPA3245 performance results are shown to make the document easier to
read. The rest of the amplifier results can be found in the appendix.
5 Output Noise
Output noise is almost cut in half for systems with PFFB implemented. The lower the noise floor, the
better users will be able to hear the small details in the audio. Table 3 lists the A-weighted noise.
Table 3. Noise – A Weighted

EVM Configuration (PVDD = 30 V, (PVDD = 30 V, (PVDD = 36 V, (PVDD = 36 V, (PVDD = 51 V,
Standard
54.5 µV 54.4 µV 62.7 µV 61.4 µV 81.5 µV
configuration
PFFB 29.5 µV 28.7 µV 30.5 µV 28.3 µV 46.2 µV
6 SNR and DNR

Signal to Noise Ratio (SNR) and Dynamic Range Ratio (DNR) are both very critical numbers to evaluate
the audio performance of an amplifier. In a single number, they summarize how the low output noise
compares to the how loud the amplifier can be. The lower the number, the better the system sounds. The
difference between SNR and DNR is that for SNR, the input signal is grounded. For DNR, the input signal
is very small (–60 dB of the input signal required achieving the power level corresponding to 1% THD+N).
Both are important to look at because some circuits may be activated in an IC at very small signals that
may not be activated when the outputs are grounded.
Table 4 lists the A-weighted SNR.
Table 4. SNR – A Weighted

Standard
–111.4 dB –111.5 dB –111.3 dB –111.8 dB –111.9 dB
configuration
PFFB –116.8 dB –116.8 dB –117.4 dB –117.8 dB –116.5 dB
Table 5 lists the A-weighted DNR.
Table 5. DNR – A Weighted

Standard
–111.2 dB –111.5 dB –111.1 dB –111.6 dB –111.9 dB
configuration
PFFB –117.1 dB –116.9 dB –117.5 dB –117.6 dB –116.5 dB
www.ti.com THD+N vs Power
7 THD+N vs Power
THD+N vs Power curves can tell you the power performance of the device at a single frequency. You can
use this curve to get power numbers for the device, such as the power at 1% THD+N and 10% THD+N.
Additionally, this curve is useful to see how the amplifier performs at lower powers that are critical for how
the amplifier sounds at typical room volumes such as 1 W to 10 W. As seen in Figure 4, PFFB in this
system slightly limits the high power performance of the device because the input op amps start to
saturate at voltages high enough to drive the amplifier to high power. The lower power performance which
is very critical to how an amplifier actually sounds to a human ear is improved by roughly 5 dB with this
PFFB configuration. As an example, the TPA3245 is used below with a 4 Ω Load in BTL with a PVDD =
30 V.
10
TPA3245, 4 :, BTL, 1 kHz, PFFB
TPA3245, 4 :, BTL, 1 kHz, No PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
0.01 0.1 1 10 100 300
Figure 4. THD+N vs Power at 1 kHz
Table 6. TPA3245: 1-kHz Input

THD+N (dB)
Output Power (W)
Standard EVM PFFB Difference
0.01 –68.383 –73.669 –5.28
0.1 –78.305 –83.263 –4.958
1 –87.610 –92.844 –5.234
10 –88.768 –94.845 –6.076
1 kHz is the standard signal for THD+N curves but other frequencies are common and useful. A 100-Hz
value gives insight in how the lower frequency performance is and can be useful for woofer or subwoofer
design.
10
TPA3245, 4 :, BTL, 100 Hz, PFFB
TPA3245, 4 :, BTL, 100 Hz, No PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
1m 10m 100m 1 10 100 300
Figure 5. THD+N vs Power at 100 Hz
THD+N vs Power www.ti.com
Table 7. TPA3245: 100-Hz Input

THD+N (dB)
Output Power (W)
0.01 –68.211 –73.418 –5.208
0.1 –78.243 –83.479 –5.236
1 –88.141 –93.446 –5.304
10 –96.802 –102.021 –5.219
6.67 kHz is also useful to see because this is typically a challenging test for class-D amplifiers. 6.67 kHz is
the highest frequency in a 20-kHz band that can still show a second (13.13 kHz) and third (20 kHz)
harmonic.
10
TPA3245, 4 :, BTL, 6.67 kHz, PFFB
TPA3245, 4 :, BTL, 6.67 kHz, No PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300
Figure 6. THD+N vs Power at 6.67 kHz
Table 8. TPA3245: 6.67-kHz Input

THD+N (dB)
Output Power (W)
0.01 –68.830 –74.306 –5.476
0.1 –77.840 –82.860 –5.020
1 –82.050 –86.783 –4.733
10 –75.523 –85.115 –9.592
www.ti.com THD+N vs Frequency
8 THD+N vs Frequency
THD+N vs Frequency curves can supplement THD+N vs Power curves to see how the amplifier performs
for all frequencies for a set power. This will ensure that the amplifier sounds good for all audible
frequencies at a certain power level. Several power levels were used. The 1-W and 10-W power levels are
important because they show how the amplifier performs at lower powers that are critical for how the
amplifier sounds at typical room volumes. A 50-W power level is also taken to show high-power
performance.
100
TPA3245, 4 :, 1 W, PFFB
TPA3245, 4 :, 1 W, No PFFB
10
1
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D005
Figure 7. THD+N vs Frequency at 1 W
Table 9. TPA3245: 1 W
THD+N (dB)
Frequency (Hz)
20 –88.199 –93.202 –5.003
100 –88.284 –93.329 –5.045
1000 –87.888 –93.063 –5.176
10000 –88.355 –91.359 –3.003
15000 –90.919 –96.330 –5.411
100
TPA3245, 4 :, BTL, 10 W, PFFB
TPA3245, 4 :, BTL, 10 W, No PFFB
10
1
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D006
THD+N vs Frequency www.ti.com
Table 10. TPA3245: 10 W

THD+N (dB)
Frequency (Hz)
20 –93.255 –99.900 –6.645
100 –97.083 –102.088 –5.005
1000 –88.966 –95.103 –6.137
10000 –82.290 –90.367 –1.077
15000 –100.874 –105.783 –4.908
100
TPA3245, 4 :, BTL, 50 W, PFFB
10
1
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D007
Table 11. TPA3245: 50 W

THD+N (dB)
Frequency (Hz)
20 –86.882 –95.380 –8.498
100 –94.246 –101.191 –6.945
1000 –86.941 –93.134 –6.193
10000 –94.173 –88.351 –5.823
15000 –107.281 –110.676 –3.395
www.ti.com SMPTE IMD
9 SMPTE IMD
Intermodulation distortion (IMD) is an important element to measure on class-D amplifiers. IMD is created
when two or more audio tones beat with one another in a non-linear device to produce undesired new
tones. STMTE IMD is a technique for measuring IMD according to the SMPTE RP120-1983 standard.
This method of IMD uses a 60-Hz tone and a 7-kHz tone mixed at a 4:1 ratio. The harmonics of the 60-Hz
tone are the primary ones measured. One useful way to view the distortion levels is to look at the sum of
the harmonics vs output power. This can give you a sense of how much of the harmonics will be audible,
and if that level gets significantly worse at high power.
0
TPA3245, 4 :, BTL, PFFB
TPA3245, 4 :, BTL, No PFFB
-20
SMPTE Ratio (dB)
-40
-60
-80
-100
-120
100m 1 10 100
Power (W) D008
Figure 10. SMPTE IMD vs Power
Another useful view is to see all of the harmonics at a particular power level. Figure 11 and Figure 12
show the 1-W and 10-W SMPTE IMD. One can see that the second harmonic is very small for the
standard configuration, and even lower with PFFB. Additionally, you can see that PFFB improves most
harmonics at 1 W and 10 W.
Table 12. SMPTE Distortion Product Ratio

Frequency 60 Hz 6.76 kHz 6.82 kHz 6.88 kHz 6.94 kHz 7.00 kHz 7.06 kHz 7.12 kHz 7.18 kHz 7.24 kHz
f1 d5 d4 d3 d2 f2 d2 d3 d4 d5
20 20
10 Standard 10 Standard
PFFB PFFB
0 0
-10 -10
-20 -20
Frequency (Hz)
Frequency (Hz)
-30 -30
-40 -40
-50 -50
-60 -60
-70 -70
-80 -80
-90 -90
-100 -100
-110 -110
f1 d5 d4 d3 d2 f2 d2 d3 d4 d5 f1 d5 d4 d3 d2 f2 d2 d3 d4 d5
D007 D010
Figure 11. SMPTE IMD Distortion Product Ratio at 1 W Figure 12. SMPTE IMD Distortion Product Ratio at 10 W
CCIF IMD www.ti.com
10 CCIF IMD
CCIF IMD is more sensitive to high-frequency nonlinearity; it mixes a 19-kHz and 20-kHz wave. CCIF IMD
versus power is a way for to measure the high-frequency nonlinearity of the device across power.
0
d2, TPA3245, Standard Configuration
d2, TPA3245, PFFB
-20 d2-d5, TPA3245, Standard Configuration
d2-d5, TPA3245, PFFB
-40
CCIF IMD (dB) -60
-80
-100
-120
10m 100m 1 10 100
Power (W) D001
Figure 13. CCIF IMD vs Power
CCIF Distortion Product Ratio allows us to take a look at which harmonics are present, and at what level.
The d2 product (1 kHz) is extremely important as it is easy to distinguish from the high frequency tones.
The TPA324x and TPA325x family has excellent d2 distortion, and PFFB further improves distortion
significantly below the audible level. The third and fifth harmonic (d3 and d5 respectively) are at higher
levels, but since they are so close to the signal tones, they are hard to perceive. One can see that PFFB
improves the system’s non-linearity and reduces d3 and d5 as well.
Table 13. CCIF Distortion Product Ratio

Frequency 1.00 kHz 2.00 kHz 17.00 kHz 18.00 kHz 19.00 kHz 20.00 kHz 21.00 kHz 22.00 kHz
d2 d4 d5 d3 f1 f2 d3 d5
0 0
Standard Standard
PFFB PFFB
-20 -20
-40 -40
Frequency (Hz)
Frequency (Hz)
-60 -60
-80 -80
-100 -100
-120 -120
d2 d4 d5 d3 f1 f2 d3 d5 d2 d4 d5 d3 f1 f2 d3 d5
D010 D011
Figure 14. CCIF IMD Distortion Product Ratio at 1 W Figure 15. CCIF IMD Distortion Product Ratio at 10 W
www.ti.com Stability Analysis
11 Stability Analysis
Stability analysis is important for PFFB to insure that the added outer PFFB loop does not cause amplifier
oscillations. With incorrectly selected PFFB component values, poor stability margins can cause the
amplifier to oscillate. This can cause the amplifier to shut down and behave erratically, especially near
clipping.
It is important to ensure the stability of the system for all conditions. For this reason, precautions must be
taken to ensure that the TPA324x and TPA325x device is stable for open load conditions when using
PFFB. For open load cases, the LC filter Q is extremely large. The Q factor of an LC filter will be affected
by changes in the load resistance and an open load will cause an extremely large Q. The Zobel network is
used to reduce the Q of the open load case by adding a resistance, or load, to the output. This reduction
in the Q factor will reduce the output ringing. See Figure 16.
Figure 16. LC Filter Open Load Response – TINA Spice Simulation
By adding more capacitance to the Zobel capacitor, the response could be further improved, but there is a
critical tradeoff. This RC network will be attached to the outputs of a high power amplifier. Therefore we
must be mindful of the voltage across the capacitor and the current through the resistor. A calculation is
done for the power dissipated in the Zobel resistor. The worst case here is a high frequency full scale
signal. The audio band goes to 20 kHz, so that will be the worst case frequency. The output terminal to
GND worst case is a full scale signal, thus PVDD is the worst case amplitude.
Table 14. Worst-Case Power Dissipation in Zobel Resistor

TPA3244 (PVDD=31.5, TPA3245 (PVDD=31.5, TPA3250 (PVDD=36 V, TPA3251 (PVDD=36 V, TPA3255 (PVDD=51V ,
Fpwm=450 kHz) Fpwm=600 kHz) Fpwm=450kHz) Fpwm=600 kHz) Fpwm=450 kHz)
P(R_z) (W) P(R_z) (W) P(R_z) (W) P(R_z) (W) P(R_z) (W)
0.095 0.095 0.124 0.124 0.273
By setting the Zobel capacitor to 220 nF, the worst case power dissipation is roughly ¼ W for the
TPA3255, and 1/8 W for the TPA3250, TPA3251, TPA3244, and TPA3245. This allows users to use
small, inexpensive components for the Zobel network.
The relationship between the Zobel capacitor value and the power dissipated for worst case is not linear.
For example, for the TPA3255 case, if the Zobel capacitor was increased to 440 nF, the power dissipated
in the resistor would be 1.09 W for worst case. If the Zobel capacitor was increased to 660 nF, the power
dissipated would be 2.44 W.
For actual audio, there is less energy in higher frequencies. A 20-kHz full scale sine wave is a very
rigorous test. For most audio applications, the Zobel resistor will dissipate very little power.
Stability Testing www.ti.com
12 Stability Testing
There are a few tests that should be performed on any PFFB configuration:
• Test overshoot for a square wave input
• Test frequency response at full scale
13 Overshoot for Square Wave Input

The TPA324x and TPA325x amplifier family has an integrated feedback loop for noise suppression, which
makes frequency domain gain and phase analysis for PFFB stability nearly impossible. Further complexity
is added since the PFFB loop includes the phase and amplitude characteristics of the output LC filter
where the internal feedback loop does not. For this reason, time domain overshoot analysis is the best
means for assessing stability.
Using the TPA324x and TPA325x EVMs setup in BTL PFFB outlined in Section 1, a 1-kHz square wave
signal was input to the system. The amplifier output was monitored on an oscilloscope to capture the
amount of overshoot from the rising edge of the input square wave. The amplifier output voltage of the
square wave signal should be large enough for good resolution with the oscilloscope used for viewing the
overshoot. However, the amplitude must not be large enough to approach clipping of the amplifier. The
nonlinearity of clipping will give inaccurate results. For this test, a 3.3 VRMS differential signal is input to the
system.
Some oscilloscopes have the built in capability to measure overshoot. Overshoot can also be calculated
using Equation 4.
Overshoot (%) = [(V_peak) – (V_ideal)] / [(V_ideal) – (V_ss)] (4)
www.ti.com Calculating Phase Margin
14 Calculating Phase Margin

After the overshoot percentage has been captured, the phase margin can be found. There are several
methods to complete this. Users can use the Stability section of the Analog Engineer's Calculator.
Figure 17. Phase Margin vs Overshoot Calculator of Analog Engineer
Alternatively, the curve in Figure 18 can be used.
Figure 18. Phase Margin Percent Overshoot Curve
TPA3245 PFFB Stability www.ti.com
The curve in Figure 18 was generated with the following code:

x = .27:.01:100;
pm = 90-(180/pi)*atan(x);
q=sqrt(x.*sqrt((1+x.^2)));
os=100*exp(-pi./(sqrt(4*q.^2-1)));
plot(pm,os)
title('Phase Margin Percent Overshoot Curve');
xlabel('Phase Margin (degrees)');
ylabel('Percent Overshoot');
grid;
The 1-kHz square wave test should be completed for possible loads the system will be subject to and an
open load test. The goal is to evaluate the stability of the loop and determine if any oscillations are
possible. By looking at a square wave, we can see the step response and determine how quickly the
oscillations decay. If the oscillations do not decay quickly enough, the output could oscillate which could
damage the device.
To complete the 1-kHz square wave test, first start with the worst case: open load. For a Class-D amplifier
in PFFB, the most unstable condition is when the amplifier output is unloaded. Without a load, the Q of the
LC filter will be extremely large. This quality factor extreme amplitude peaking at a frequency determined
by the component values of the inductor (L) and capacitor (C). Furthermore, the higher the Q, the quicker
phase will shift –180° for incremental increase in frequency, meaning the amplitude of the filter will still be
very large when large phase shifts occurs; this causes instability.
If the amplifier is proven to have a stable open load, then the likelihood of stability issues when the
amplifier output is loaded are reduced.
15 TPA3245 PFFB Stability

Using an oscilloscope, measure the signal after the LC filter with no load connected to the output. Supply
a 1-kHz square wave signal to the amplifier input. Set the amplitude either to the maximum amplitude
signal your system will supply or to the highest input that does not produce clipping, whichever is lower.
For example, a 3.3-VRMS signal 1-kHz square wave signal is supplied to the TPA3245EVM with PFFB
components installed before the output started clipping. Figure 19 shows the output wave.
Figure 19. TPA3245 PFFB 1-kHz Square Wave Open Response
The scope images have an overshoot of 73%, and a phase margin of 11°.
www.ti.com TPA3245 PFFB Stability
The 11° is rather small, but this is a worst case test. It is also important to note how quickly the oscillations
decay. This very quick decay points to good stability for the loop. Figure 20 shows the open load response
to the TPA3245 without PFFB.
Figure 20. TPA3245 Standard Configuration 1-kHz Square Wave Open Response
The 1-kHz square wave then should be tested on all other loads the system is to support. Ensure that the
load you are using is able to support the power the amplifier will be supplying. For the TPA3245, this is
about 115 W for 4 Ω and 60 W for 8 Ω.
Figure 21. TPA3245 PFFB 1-kHz Square Wave 4-Ω Response
With a 4-Ω load, the output has an 18.4% overshoot and a phase margin of 50°.
Figure 22. TPA3245 PFFB 1-kHz Square Wave 8-Ω Response

Frequency Reponse at Full Scale www.ti.com
The square wave tests for the TPA3250, TPA3251, TPA3245, TPA3244, and TPA3255 can be found in
Appendix A, Appendix B, Appendix C, Appendix D, and Appendix E.
16 Frequency Reponse at Full Scale

The frequency response of the device should also be tested to ensure that no signals inside the audio
band can cause instability for the device. This should be done for open load, and all loads that the system
is to support. This test should be done at the highest output voltage that the device is to support.
2
1.5
1
Output Level (dB)
0.5
-0.5
-1
TPA3245, PFFB, Open Load Response
-1.5 TPA3245, PFFB, 4 : Response
TPA3245, PFFB, 8 : Response
-2
20 100 1k 10k 20k
Frequency (Hz) D002
Figure 23. PFFB Audible Frequency Response
If a fault occurs during either one of these tests there is an issue with stability.
It is important to note this should be done for all frequencies this device is expected to pass. If, for some
reason the amplifier will receive inputs outside of the audio range at high levels, this should be tested as
well. The system is more likely to have stability issues at higher frequencies, especially for open load, due
to the peaking of the LC filter. Figure 24 shows the extended PFFB frequency response.
15
TPA3245, PFFB, Open Load Response
TPA3245, PFFB, 4 : Response
10 TPA3245, PFFB, 8 : Response
Output Level (dB)
-5
-10
-15
20 100 1k 10k 100k
Frequency (Hz) D015
Figure 24. Extended PFFB Frequency Response
If the device passes the overshoot for a square wave input and the frequency response at full scale for all
conditions of the audio system (PVDD voltage, loading conditions, temperature, and more), the system is
considered stable.
Appendix A
TPA3244
A.1 TPA3244 EVM PFFB Test Results
Table 15. TPA3244 Summary

Parameter Standard PFFB
Gain (dB) 18 11.8
Negative feedback (dB) 0 6.2
Output noise (µV) A-weighted 54.5 29.5
SNR (dB) A-weighted –111.4 –116.8
DNR (dB) A-weighted –111.2 –117.1
A.2 TPA3244 THD+N vs Power
10
1
THD + N ( )
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300
Table 16. TPA3244: 1 kHz

Output Power (W) THD+N (dB)
0.01 –68.196 –73.677 –5.481
0.1 –78.298 –83.684 –5.386
1 –87.277 –92.770 –5.492
10 –87.343 –93.570 –6.226
TPA3244 THD+N vs Power www.ti.com
10
TPA3244, 4 :, BTL, 100 Hz, PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300
Table 17. TPA3244: 100 Hz

0.01 –68.296 –73.695 –5.399
0.1 –78.380 –83.605 –5.226
1 –88.286 –93.599 –5.313
10 –96.775 –102.759 –5.985
10
TPA3244, 4 :, BTL, 6.67 kHz, PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300
Table 18. TPA3244: 6.67 kHz

0.01 –69.063 –74.142 –5.079
0.1 –77.034 –81.750 –4.717
1 –79.047 –83.735 –4.688
10 –74.878 –81.555 –6.677
www.ti.com TPA3244 THD+N vs Frequency
A.3 TPA3244 THD+N vs Frequency
100
TPA3244, 4 :, BTL, 1 W, PFFB
10
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D019

Frequency (Hz) THD+N (dB)
20 –88.202 –93.291 –5.089
100 –88.468 –93.554 –5.086
1000 –87.740 –92.893 –5.152
10000 –86.427 –87.280 –0.854
15000 –90.679 –96.657 –5.977
100
TPA3244, 4 :, BTL, 10 W, PFFB
10
1
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D020
Table 20. TPA3244: 10 W

20 –93.200 –99.457 –6.257
100 –97.002 –102.704 –5.702
1000 –88.101 –94.294 –6.193
10000 –90.901 –93.418 –2.517
15000 –101.101 –106.543 –5.442
TPA3244 SMPTE Distortion Product Ratio www.ti.com
100
TPA3244, 4 :, BTL, 50 W, PFFB
10
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D021
Table 21. TPA3244: 50 W

20 –88.218 –95.958 –7.740
100 –93.263 –101.948 –8.685
1000 –86.504 –92.282 –5.779
10000 –89.209 –101.776 –12.567
15000 –107.020 –111.545 –4.525
A.4 TPA3244 SMPTE Distortion Product Ratio
30 30
Standard Standard
10 PFFB 10 PFFB
-10 -10
Frequency (Hz)
Frequency (Hz)
-30 -30
-50 -50
-70 -70
-90 -90
-110 -110
D022 D023
Figure 31. SMPTE IMD – TPA3244 at 1 W Figure 32. SMPTE IMD – TPA3244 at 10 W
www.ti.com TPA3244 CCIF Distortion Product Ratio
20
0 TPA3244, 4 :, BTL, No PFFB
-20
SMPTE Ratio (dB)

-40
-60
-80
-100
-120
100m 1 10 100
Power (W) D024
Figure 33. SMPTE Ratio vs Output Power
A.5 TPA3244 CCIF Distortion Product Ratio
0 0
-10 Standard Standard
PFFB PFFB
-20 -20
-30
-40 -40
Frequency (Hz)
Frequency (Hz)
-50
-60 -60
-70
-80 -80
-90
-100 -100
-110
-120 -120
D025 D026
Figure 34. CCIF IMD – TPA3244 at 1 W Figure 35. CCIF IMD – TPA3244 at 10 W
0
d2, TPA3244, PFFB
-40
CCIF IMD (dB)
-60
-80
-100
-120
100m 1 10 100
Power (W) D027
Figure 36. CCIF Response
TPA3244 Stability Analysis www.ti.com
A.6 TPA3244 Stability Analysis
Table 22. TPA3244 – PFFB

Load (Ω) Overshoot (%) Phase Margin (degrees)
OL 77.9 9.1
4 19.1 49.0
8 36.3 34.0
Figure 37. TPA3244 PFFB – Open-Load Response Figure 38. TPA3244 PFFB – 4-Ω Response
Figure 39. TPA3244 PFFB – 8-Ω Response
Appendix B
TPA3245
B.1 TPA3245 EVM PFFB Test Results

Gain (dB) 18 11.8
B.2 THD+N vs Power

10
1
THD + N (%)
0.1
0.01
0.001
0.0001
0.01 0.1 1 10 100 300

THD+N (dB)
Output Power
0.01 –68.383 –73.669 –5.28
0.1 –78.305 –83.263 –4.958
1 –87.610 –92.844 –5.234
10 –88.768 –94.845 –6.076
THD+N vs Power www.ti.com
10
TPA3245, 4 :, BTL, 100 Hz, PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
1m 10m 100m 1 10 100 300

THD+N (dB)
Output Power
0.01 –68.211 –73.418 –5.208
0.1 –78.243 –83.479 –5.236
1 –88.141 –93.446 –5.304
10 –96.802 –102.021 –5.219
10
TPA3245, 4 :, BTL, 6.67 kHz, PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300

THD+N (dB)
Output Power
0.01 –68.830 –74.306 –5.476
0.1 –77.840 –82.860 –5.020
1 –82.050 –86.783 –4.733
10 –75.523 –85.115 –9.592
B.3 THD+N vs Frequency

100
TPA3245, 4 :, 1 W, PFFB
TPA3245, 4 :, 1 W, No PFFB
10
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D005

THD+N (dB)
Frequency
20 –88.199 –93.202 –5.003
100 –88.284 –93.329 –5.045
1000 –87.888 –93.063 –5.176
10000 –88.355 –91.359 –3.003
15000 –90.919 –96.330 –5.411
100
TPA3245, 4 :, BTL, 10 W, PFFB
10
1
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D006
Table 28. TPA3245: 10 W

THD+N (dB)
Frequency
20 –93.255 –99.900 –6.645
100 –97.083 –102.088 –5.005
1000 –88.966 –95.103 –6.137
10000 –82.290 –90.367 –1.077
15000 –100.874 –105.783 –4.908
TPA3245 – SMPTE Distortion Product Ratio www.ti.com
100
TPA3245, 4 :, BTL, 50 W, PFFB
10
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D007
Table 29. TPA3245: 50 W

THD+N (dB)
Frequency
20 –86.882 –95.380 –8.498
100 –94.246 –101.191 –6.945
1000 –86.941 –93.134 –6.193
10000 –94.173 –88.351 –5.823
15000 –107.281 –110.676 –3.395
B.4 TPA3245 – SMPTE Distortion Product Ratio
20 20
10 Standard 10 Standard
PFFB PFFB
0 0
-10 -10
-20 -20
Frequency (Hz)
Frequency (Hz)
-30 -30
-40 -40
-50 -50
-60 -60
-70 -70
-80 -80
-90 -90
-100 -100
-110 -110
D007 D010
www.ti.com TPA3245 – CCIF Distortion Product Ratio
0
-20
SMPTE Ratio (dB)

-40
-60
-80
-100
-120
100m 1 10 100
Power (W) D008
B.5 TPA3245 – CCIF Distortion Product Ratio
0 0
Standard Standard
PFFB PFFB
-20 -20
-40 -40
Frequency (Hz)
Frequency (Hz)
-60 -60
-80 -80
-100 -100
-120 -120
D010 D011
0
d2, TPA3245, PFFB
-40
CCIF IMD (dB)
-60
-80
-100
-120
10m 100m 1 10 100
Power (W) D001
B.6 TPA3245 Stability Analysis

OL 73.0 11.4
4 18.1 50.0
8 35.0 35.0
Appendix C
TPA3250
C.1 TPA3250 EVM PFFB Test Results

Gain (dB) 20 12.75
C.2 TPA3250 THD+N vs Power

10
1
THD + N (%)
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300

0.01 –66.926 –72.981 –6.055
0.1 –76.911 –83.258 –6.348
1 –86.555 –92.542 –5.987
10 –87.940 –94.371 –6.432
10
TPA3250, 4 :, BTL, 100 Hz, PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
1m 10m 100m 1 10 100 300

THD+N (dB)
Output Power (W) Standard EVM PFFB Difference
0.01 –66.918 –72.774 –5.856
0.1 –76.941 –82.786 –5.845
1 –86.885 –92.682 –5.797
10 –95.880 –101.513 –5.633
10
TPA3250, 4 :, BTL, 6.67 kHz, PFFB
TPA3250, 4 :, BTL, 6.67 kHz, PFFB
1
THD + N (%)
0.1
0.01
0.001
10m 100m 1 10 100 300

THD+N (dB)
0.01 –67.796 –73.258 –5.462
0.1 –77.577 –83.145 –5.568
1 –82.605 –87.324 –4.719
10 –76.296 –81.575 –5.279
www.ti.com TPA3250 THD+N vs Frequency
C.3 TPA3250 THD+N vs Frequency

100
TPA3250, 4 :, BTL, 1 W, PFFB
10
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D031

THD+N (dB)
Frequency (Hz) Standard EVM PFFB Difference
20 –86.917 –92.803 –5.885
100 –86.898 –92.881 –5.983
1000 –86.771 –92.529 –5.759
10000 –83.247 –91.568 –8.321
15000 –89.403 –95.218 –5.815
100
TPA3250, 4 :, BTL, 10 W, PFFB
10
1
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D032
Table 36. TPA3250: 10 W

THD+N (dB)
20 –96.172 –101.561 –5.388
100 –95.899 –101.641 –5.742
1000 –88.297 –95.230 –6.933
10000 –84.369 –87.426 –3.057
15000 –99.568 –105.261 –5.693
100
TPA3250, 4 :, BTL, 50 W, PFFB
10
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D033
Table 37. TPA3245: 50 W

THD+N (dB)
20 –96.174 –95.165 1.009
100 –94.061 –96.871 –2.810
1000 –86.436 –94.864 –8.428
10000 –83.185 –87.548 –4.363
15000 –106.203 –110.651 –4.448
C.4 TPA3250 – SMPTE Distortion Product Ratio
30 30
Standard Standard
10 PFFB 10 PFFB
-10 -10
Frequency (Hz)
Frequency (Hz)
-30 -30
-50 -50
-70 -70
-90 -90
-110 -110
D034 D035
0
-20
SMPTE Ratio (dB)

-40
-60
-80
-100
-120
0.1 1 10 100
Power (W) D036
C.5 TPA3250 – CCIF Distortion Product Ratio
0 0
Standard -10 Standard
PFFB PFFB
-20 -20
-30
-40 -40
Frequency (Hz)
Frequency (Hz)
-50
-60 -60
-70
-80 -80
-90
-100 -100
-110
-120 -120
D037 D038
0
d2, TPA3250, PFFB
-40
CCIF IMD (dB)
-60
-80
-100
-120
0.1 1 10 100
Power (W) D039
C.6 TPA3250 Stability Analysis

OL 51.0 23.7
4 18.4 49.7
8 29.5 39.5
Appendix D
TPA3251
D.1 TPA3251 EVM PFFB Test Results

Gain (dB) 20 12.75
D.2 TPA3251 THD+N vs Power

10
1
THD + N (%)
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300

THD+N (dB)
0.01 –67.339 –73.622 –6.283
0.1 –77.261 –83.708 –6.447
1 –87.187 –93.660 –6.473
10 –91.852 –95.140 –3.288
10
TPA3251, 4 :, BTL, 100 Hz, PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300

THD+N (dB)
0.01 –67.305 –73.801 –6.496
0.1 –77.372 –83.615 –6.243
1 –87.228 –93.610 –6.382
10 –96.038 –101.894 –5.856
10
TPA3251, 4 :, BTL, 6.67 kHz, PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300

THD+N (dB)
0.01 –68.200 –74.304 –6.104
0.1 –77.788 –83.742 –5.954
1 –85.894 –90.153 –4.258
10 –77.789 –81.460 –3.670
D.3 THD+N vs Frequency

100
TPA3251, 4 :, BTL, 1 W, PFFB
10
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D043

THD+N (dB)
20 –87.132 –93.440 –6.308
100 –87.398 –93.682 –6.284
1000 –87.331 –93.405 –6.074
10000 –88.003 –93.690 –5.687
15000 –89.834 –95.767 –5.933
100
TPA3251, 4 :, BTL, 10 W, PFFB
10
1
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D044
Table 44. TPA3251: 10 W

THD+N (dB)
20 –92.789 –99.585 –6.796
100 –96.070 –102.051 –5.981
1000 –91.882 –96.705 –4.823
10000 –98.215 –95.809 2.406
15000 –99.487 –106.007 –6.520
100
TPA3251, 4 :, BTL, 50 W, PFFB
10
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D045
Table 45. TPA3251: 50 W

THD+N (dB)
20 –85.180 –92.110 –6.930
100 –93.986 –95.960 –1.974
1000 –90.983 –94.627 –3.644
10000 –93.689 –95.777 –2.088
15000 –106.407 –111.074 –4.667
D.4 TPA3251 – SMPTE Distortion Product Ratio
30 30
Standard Standard
10 PFFB 10 PFFB
-10 -10
Frequency (Hz)
Frequency (Hz)
-30 -30
-50 -50
-70 -70
-90 -90
-110 -110
f1 d5 d4 d3 d2 d2 d3 d3 d4 d5 f1 d5 d4 d3 d2 f2 d2 d3 d4 d5
D046 D047
0
-20
SMPTE Ratio (dB)

-40
-60
-80
-100
-120
0.1 1 10 100
Power (W) D048
D.5 TPA3251 – CCIF Distortion Product Ratio
0 0
Standard Standard
PFFB PFFB
-20 -20
-40 -40
Frequency (Hz)
Frequency (Hz)
-60 -60
-80 -80
-100 -100
-120 -120
D049 D050
0
d2, TPA3251, PFFB
-40
CCIF IMD (dB)
-60
-80
-100
-120
100m 1 10 100
Power (W) D051
D.6 TPA3251 Stability Analysis

OL 59.8 18.3
4 18.4 49.7
8 29.5 39.5
Appendix E
TPA3255
E.1 TPA3255 EVM PFFB Test Results

Gain (dB) 21.5 15.9
E.2 TPA3255 THD+N vs Power

10
1
THD + N (%)
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300

0.01 –64.882 –69.511 –4.630
0.1 –74.564 –78.990 –4.426
1 –84.682 –89.125 –4.443
10 –88.905 –93.545 –4.640
100 –85.619 –89.501 –3.882
10
TPA3255, 4 :, BTL, 100 Hz, PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300

0.01 –64.597 –69.091 –4.494
0.1 –74.605 –79.294 –4.690
1 –84.516 –89.081 –4.564
10 –93.713 –98.335 –4.622
100 –93.498 –97.274 –3.776
10
TPA3255, 4 :, BTL, 6.67 kHz, PFFB
1
THD + N (%)
0.1
0.01
0.001
0.0001
10m 100m 1 10 100 300

0.01 –65.289 –69.784 –4.495
0.1 –75.323 –79.352 –4.029
1 –74.158 –74.714 –0.556
10 –77.214 –82.780 –5.565
100 –68.772 –72.664 –3.892
E.3 THD+N vs Frequency

100
TPA3255, 4 :, BTL, 1 W, PFFB
10
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D055

20 –84.620 –88.456 –3.836
100 –84.644 –89.132 –4.488
1000 –84.682 –89.176 –4.494
10000 –78.621 –89.730 –11.109
15000 –87.372 –92.690 –5.319
100
TPA3255, 4 :, BTL, 10 W, PFFB
10
1
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D056
Table 52. TPA3255: 10 W

20 –91.195 –95.764 –4.569
100 –93.794 –98.257 –4.464
1000 –89.108 –93.560 –4.452
10000 –88.629 –92.993 –4.364
15000 –97.007 –102.332 –5.324
100
TPA3255, 4 :, BTL, 50 W, PFFB
10
THD + N (%)
0.1
0.01
0.001
0.0001
20 100 1k 10k 20k
Frequency (Hz) D056
Table 53. TPA3255: 50 W

20 –85.610 –92.025 –6.415
100 –95.919 –99.465 –3.545
1000 –87.237 –91.221 –3.984
10000 –96.057 –93.741 2.316
15000 –104.838 –108.176 –3.338
E.4 TPA3255 – SMPTE Distortion Product Ratio
30 20
Standard Standard
10 PFFB 0 PFFB
-10 -20
Frequency (Hz)
Frequency (Hz)
-30 -40
-50 -60
-70 -80
-90 -100
-110 -120
D058 D059
0
-20
SMPTE Ratio (dB)

-40
-60
-80
-100
-120
100m 1 10 100
Power (W) D060
E.5 TPA3245 – CCIF Distortion Product Ratio
0 0
Standard Standard
PFFB PFFB
-20 -20
-40 -40
Frequency (Hz)
Frequency (Hz)
-60 -60
-80 -80
-100 -100
-120 -120
D061 D062
0
d2, TPA3255, PFFB
-40
CCIF IMD (dB)
-60
-80
-100
-120
100m 1 10 100
Power (W) D063
E.6 TPA3255 Stability Analysis

OL 78.4 8.8
4 16.3 51.8
8 35.0 35.0
www.ti.com Revision History
Revision History
NOTE: Page numbers for previous revisions may differ from page numbers in the current version.
Changes from Original (September 2017) to A Revision ............................................................................................... Page
• Changed TPA3244 and TPA3245 column from "PVDD = 31.5" to "PVDD = 30 V" in Table 1, Table 2, Table 3, Table 4,
and Table 5. ................................................................................................................................ 4
• Changed TPA3244 column from "Fpwm = kHz" to "Fpwm = 450 kHz" in Table 1, Table 2, Table 3, Table 4, and
Table 5. ..................................................................................................................................... 4
• Changed C_out values from µH to µF .................................................................................................. 4
• Changed TND+N to THD+N in THD+N vs Power Inductor Comparison .......................................................... 5
• Corrected THD+N vs Power Inductor Comparison legend .......................................................................... 5
• Changed "with a PVDD = 31.5 V" to "with a PVDD = 30 V" in Section 7 .......................................................... 7
• Switched "Standard EVM" and "PFFB" values in Table 6, Table 7, Table 8, Table 9, Table 10, Table 11, Table 16,
Table 17, Table 18, Table 19, Table 20, Table 21, Table 24, Table 25, Table 26, Table 27, Table 28, Table 29, Table 32,
Table 33, Table 34, Table 35, Table 36, Table 37, Table 40, Table 41, Table 42, Table 43, Table 44, Table 45, Table 48,
Table 49, Table 50, Table 51, Table 52, and Table 53. .............................................................................. 7
SLAA788A – September 2017 – Revised March 2018 Revision History 49

IMPORTANT NOTICE FOR TI DESIGN INFORMATION AND RESOURCES
Texas Instruments Incorporated (‘TI”) technical, application or other design advice, services or information, including, but not limited to,
reference designs and materials relating to evaluation modules, (collectively, “TI Resources”) are intended to assist designers who are
developing applications that incorporate TI products; by downloading, accessing or using any particular TI Resource in any way, you
(individually or, if you are acting on behalf of a company, your company) agree to use it solely for this purpose and subject to the terms of
this Notice.
TI’s provision of TI Resources does not expand or otherwise alter TI’s applicable published warranties or warranty disclaimers for TI
products, and no additional obligations or liabilities arise from TI providing such TI Resources. TI reserves the right to make corrections,
enhancements, improvements and other changes to its TI Resources.
You understand and agree that you remain responsible for using your independent analysis, evaluation and judgment in designing your
applications and that you have full and exclusive responsibility to assure the safety of your applications and compliance of your applications
(and of all TI products used in or for your applications) with all applicable regulations, laws and other applicable requirements. You
represent that, with respect to your applications, you have all the necessary expertise to create and implement safeguards that (1)
anticipate dangerous consequences of failures, (2) monitor failures and their consequences, and (3) lessen the likelihood of failures that
might cause harm and take appropriate actions. You agree that prior to using or distributing any applications that include TI products, you
will thoroughly test such applications and the functionality of such TI products as used in such applications. TI has not conducted any
testing other than that specifically described in the published documentation for a particular TI Resource.
You are authorized to use, copy and modify any individual TI Resource only in connection with the development of applications that include
the TI product(s) identified in such TI Resource. NO OTHER LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE TO
ANY OTHER TI INTELLECTUAL PROPERTY RIGHT, AND NO LICENSE TO ANY TECHNOLOGY OR INTELLECTUAL PROPERTY
RIGHT OF TI OR ANY THIRD PARTY IS GRANTED HEREIN, including but not limited to any patent right, copyright, mask work right, or
other intellectual property right relating to any combination, machine, or process in which TI products or services are used. Information
regarding or referencing third-party products or services does not constitute a license to use such products or services, or a warranty or
endorsement thereof. Use of TI Resources 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.
TI RESOURCES ARE PROVIDED “AS IS” AND WITH ALL FAULTS. TI DISCLAIMS ALL OTHER WARRANTIES OR
REPRESENTATIONS, EXPRESS OR IMPLIED, REGARDING TI RESOURCES OR USE THEREOF, INCLUDING BUT NOT LIMITED TO
ACCURACY OR COMPLETENESS, TITLE, ANY EPIDEMIC FAILURE WARRANTY AND ANY IMPLIED WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NON-INFRINGEMENT OF ANY THIRD PARTY INTELLECTUAL
PROPERTY RIGHTS.
TI SHALL NOT BE LIABLE FOR AND SHALL NOT DEFEND OR INDEMNIFY YOU AGAINST ANY CLAIM, INCLUDING BUT NOT
LIMITED TO ANY INFRINGEMENT CLAIM THAT RELATES TO OR IS BASED ON ANY COMBINATION OF PRODUCTS EVEN IF
DESCRIBED IN TI RESOURCES OR OTHERWISE. IN NO EVENT SHALL TI BE LIABLE FOR ANY ACTUAL, DIRECT, SPECIAL,
COLLATERAL, INDIRECT, PUNITIVE, INCIDENTAL, CONSEQUENTIAL OR EXEMPLARY DAMAGES IN CONNECTION WITH OR
ARISING OUT OF TI RESOURCES OR USE THEREOF, AND REGARDLESS OF WHETHER TI HAS BEEN ADVISED OF THE
POSSIBILITY OF SUCH DAMAGES.
You agree to fully indemnify TI and its representatives against any damages, costs, losses, and/or liabilities arising out of your non-
compliance with the terms and provisions of this Notice.
This Notice applies to TI Resources. Additional terms apply to the use and purchase of certain types of materials, TI products and services.
These include; without limitation, TI’s standard terms for semiconductor products http://www.ti.com/sc/docs/stdterms.htm), evaluation
modules, and samples (http://www.ti.com/sc/docs/sampterms.htm).
Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265

TPA324x and TPA325x Post-Filter Feedback

Uploaded by

Copyright:

Available Formats

TPA324x and TPA325x Post-Filter Feedback

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

TPA324x and TPA325x Post-Filter Feedback

Uploaded by

Copyright:

Available Formats

Application Report

SLAA788A – September 2017 – Revised March 2018

TPA324x and TPA325x Post-Filter Feedback

Dan Kisling, Matthew Beardsworth

Figure 1. Post-Filter Feedback Loop

The following components are required for PFFB implementation:

R_op-fb C_z R_z

Figure 2. Passive PFFB Implementation

Table 1. Recommended PFFB Component Values

2 Closed Loop Gain

Table 2. PFFB Parameters

Figure 3. THD+N vs Power Inductor Comparison

Table 3. Noise – A Weighted

6 SNR and DNR

Table 4. SNR – A Weighted

Table 5 lists the A-weighted DNR.

Table 5. DNR – A Weighted

Figure 4. THD+N vs Power at 1 kHz

Table 6. TPA3245: 1-kHz Input

Figure 5. THD+N vs Power at 100 Hz

Table 7. TPA3245: 100-Hz Input

Figure 6. THD+N vs Power at 6.67 kHz

Table 8. TPA3245: 6.67-kHz Input

Figure 7. THD+N vs Frequency at 1 W

Figure 8. THD+N vs Frequency at 10 W

Table 10. TPA3245: 10 W

Figure 9. THD+N vs Frequency at 50 W

Table 11. TPA3245: 50 W

Figure 10. SMPTE IMD vs Power

Table 12. SMPTE Distortion Product Ratio

CCIF IMD (dB) -60

Figure 13. CCIF IMD vs Power

Table 13. CCIF Distortion Product Ratio

Figure 16. LC Filter Open Load Response – TINA Spice Simulation

Table 14. Worst-Case Power Dissipation in Zobel Resistor

13 Overshoot for Square Wave Input

14 Calculating Phase Margin

Figure 17. Phase Margin vs Overshoot Calculator of Analog Engineer

Alternatively, the curve in Figure 18 can be used.

Figure 18. Phase Margin Percent Overshoot Curve

The curve in Figure 18 was generated with the following code:

15 TPA3245 PFFB Stability

Figure 19. TPA3245 PFFB 1-kHz Square Wave Open Response

Figure 21. TPA3245 PFFB 1-kHz Square Wave 4-Ω Response

Figure 22. TPA3245 PFFB 1-kHz Square Wave 8-Ω Response

16 Frequency Reponse at Full Scale

Figure 23. PFFB Audible Frequency Response

Figure 24. Extended PFFB Frequency Response

A.1 TPA3244 EVM PFFB Test Results

Table 15. TPA3244 Summary

A.2 TPA3244 THD+N vs Power

Figure 25. THD+N vs Power at 1 kHz

Table 16. TPA3244: 1 kHz

Figure 26. THD+N vs Power at 100 Hz

Table 17. TPA3244: 100 Hz

Figure 27. THD+N vs Power at 6.67 kHz

Table 18. TPA3244: 6.67 kHz

A.3 TPA3244 THD+N vs Frequency

Figure 28. THD+N vs Frequency at 1 W

Table 19. TPA3244: 1 W