Design

Mic Pre-Amplifier

The designed circuit is a microphone preamp with two channels, compatible of both balanced and unbalanced input signal, and is able to boost the signal level from input MIC signal level of -60 ~ -40 dB to output LINE signal level of -20 ~ 0 dB with an adjustable gain from 20 ~ 40dB.

Preamp Design
Figure 1.1. Preamp Design

  • Input Selection Stage:
    DPDT switch is applied to switch the input between balanced and unbalanced signal connectors.

  • Noise Cancelling Stage:
    INA103KP low noise, low distortion instrumentation amplifier is used with its output stage gain adjust mode, to cancel out the noise from the balanced signal and amplifier the signal by 20dB at the same time.

    INA103 Gain Adjustment of Output Stage
    Figure 1.2. INA103 Gain Adjustment of Output Stage

    \[A_{V}=\frac{(R_2 || 12 k\omega)+R_1+R_3}{R_2 || 12 k\omega}\]

    [12 \(k\omega\) as internal resistance between 10 and 11]

    Gain \(A_V\) set to 10 (20dB) due to design purpose.
    \(R_1\) and \(R_3\) set to 1.2 \(k\omega\), \(R_2\) set to 273 \(\omega\) to meet main requirement as well as limiting current

  • Band-pass Filtering Stage:
    Active band-pass filter with cutoff frequency 20Hz and 20kHz, using TL082 operational amplifier, is implemented to filter out the noises with frequency out of human hearing range.

    Active Band-pass Filter using TL082 (20Hz to 20kHz)
    Figure 1.3. Active Band-pass Filter using TL082 (20Hz to 20kHz)

    \[f_{c1}=\frac{1}{2\pi R_1C_1}=\frac{1}{2\pi \times 10 k\omega \times 0.8\mu F}=19.89 Hz\] \[f_{c2}=\frac{1}{2\pi R_2C_2}=\frac{1}{2\pi \times 100 k\omega \times 80 pF}=19.89 kHz\]
  • Output Adjustment Stage:
    TL082 operational amplifier is used to invert the signal that is inverted during the band-pass filtering stage to create an output signal in phase with the input signal. A 10k potentiometer is also implemented to enable gain adjust.

    Output Adjustment Amplifier using TL082
    Figure 1.4. Output Adjustment Amplifier using TL082

    \[A_V=\frac{R_{10}+R_{11}}{R_9}=\frac{(0\omega \sim 10k\omega)+1.1k\omega}{1.1 k\omega}=1 \sim 10.09=0dB \sim 20.08dB\]

5 Band Equalizer

The designed circuit is a 5-band equalizer with 5 set points across human hearing range, 108Hz, 343Hz, 1.08kHz, 3.43kHz, and 10.8kHz. Each set point has gain adjusted by external potentiometer.

KA2223 5-band graphic equalizer amplifier chip is used for the 5-band equalizer application, with resonance frequency controlled by external capacitors and gain control through external adjustable resistors.

5 Band Equalizer Design
Figure 2.1. 5 Band Equalizer Design

\[f=\frac{1}{2\pi \sqrt{R_1R_2C_1C_2}}\]

[\(R_1\),\(R_2\) are 1.2 \(k\omega\) and 68\(k\omega\) on-chip resistors]

\(f_1=\frac{1}{2\pi \sqrt{1.2k\omega \times 68 k\omega \times 0.68 \mu F \times 39 nF}}=108.19 Hz\)
\(f_2=\frac{1}{2\pi \sqrt{1.2k\omega \times 68 k\omega \times 0.22 \mu F \times 12 nF}}=342.90 Hz\)
\(f_3=\frac{1}{2\pi \sqrt{1.2k\omega \times 68 k\omega \times 68 nF \times 3.9 nF}}=1.08 kHz\)
\(f_4=\frac{1}{2\pi \sqrt{1.2k\omega \times 68 k\omega \times 22 nF \times 1.2 nF}}=3.43 kHz\)
\(f_5=\frac{1}{2\pi \sqrt{1.2k\omega \times 68 k\omega \times 6.8 nF \times 390 pF}}=10.82 kHz\)

Power Amplifier

The designed circuit is a class-AB power amplifier with gain of 31 (28.8 in real test) at 1kHz, highest average output power of 38W (25.92W in real test) across 8Ω load at 1kHz and ±28V supply voltage, and surface temperature maintained at 51.8℃ during highest power output.

Power Amplifier Schematic
Figure 3.1. Power Amplifier Schematic

  • Gain Control:
    \(R_2\), \(R_3\), and internal resistor at output pin work together to control the gain of the amplifier.

    \[Gain=1+\frac{R_3+R_{in}}{R_2}=1+\frac{20 k\omega + 10 k\omega}{1 k\omega}=31\]
  • Mute Control:
    LM3886TF chip is designed with muting function that turns off when over 0.5mA is drawn from pin 8. The mute switch is applied to toggle the mute function and the led is used to indicate the status of output.

    \[I_{drawn}=\frac{|V_{DD}|-2\times V_{diode}}{R_4}=\frac{|-28 V|-2\times 1.3V}{10 k\omega}=2.54 mA > 0.5 mA\]

Testing

Mic Pre-Amplifier

  • The unbalanced signal input is used for testing, and a sine wave signal with -40dB(10mV) RMS is applied at the input.
  • The frequency of the input signal sweeps from 1Hz to 100kHz to obtain the frequency response of the circuit.
  • SNR is also calculated through the measurements of noise and signal power.

Frequency Response

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 10 8 0.8 -1.94 dB
2 10 24 2.4 7.60 dB
4 10 80 8.0 18.06 dB
8 10 260 26.0 28.30 dB
10 10 340 34.0 30.63 dB
20 10 620 62.0 35.85 dB
40 10 840 84.0 38.49 dB
80 10 940 94.0 39.46 dB
100 10 960 96.0 39.65 dB
200 10 960 96.0 39.65 dB
400 10 960 96.0 39.65 dB
800 10 960 96.0 39.65 dB
1000 10 960 96.0 39.65 dB
2000 10 960 96.0 39.65 dB
4000 10 920 92.0 39.28 dB
8000 10 840 84.0 38.49 dB
10000 10 780 78.0 37.84 dB
20000 10 500 50.0 33.98 dB
40000 10 232 23.2 27.31 dB
80000 10 176 17.6 24.91 dB
100000 10 144 14.4 23.17 dB

Table 1.1. Frequency Response Data

\[\textrm{Ratio Gain}=\frac{\textrm{Output RMS}}{\textrm{Input RMS}}\] \[\textrm{dB Gain}=20 log_{10} \textrm{Ratio Gain}\]

Frequency Response with Ratio Gain
Figure 1.5. Frequency Response with Ratio Gain

Frequency Response with dB Gain
Figure 1.6. Frequency Response with dB Gain

SNR:

\[\textrm{Noise RMS}=14.4 mV_{pp}=5.09 mV_{rms}=-45.86 dBV\] \[\textrm{Signal RMS}=1V_{rms}=0 dBV\] \[SNR = 0dBV-(-45.86 dBV)=45.86dB\]

5 Band Equalizer

  • A sine wave signal input is used for testing, with an amplitude of 500 \(mV_{rms}\).
  • The frequency of the input signal sweeps from 1 Hz to 100 kHz to obtain the frequency response of the circuit.
  • THD of the chip is given in datasheet with value of 0.02% at 1 kHz.

Frequency Response

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 500 65 0.1 -17.72 dB
2 500 80 0.2 -15.92 dB
4 500 115.5 0.2 -12.73 dB
8 500 215 0.4 -7.33 dB
10 500 273 0.5 -5.26 dB
20 500 352 0.7 -3.05 dB
40 500 399 0.8 -1.96 dB
80 500 435 0.9 -1.21 dB
100 500 438 0.9 -1.15 dB
200 500 436 0.9 -1.19 dB
400 500 420 0.8 -1.51 dB
800 500 400 0.8 -1.94 dB
1000 500 395 0.8 -2.05 dB
2000 500 364 0.7 -2.76 dB
4000 500 274 0.5 -5.22 dB
8000 500 155 0.3 -10.17 dB
10000 500 139 0.3 -11.12 dB
20000 500 125 0.3 -12.04 dB
40000 500 120 0.2 -12.40 dB
80000 500 117 0.2 -12.62 dB
100000 500 117 0.2 -12.62 dB

Table 2.1. 108 Hz Frequency Response Data

108 Hz Frequency Response in dB
Figure 2.2. 108 Hz Set Point Circuit Frequency Response in dB

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 500 84 0.2 -15.49 dB
2 500 90 0.2 -14.89 dB
4 500 122 0.2 -12.25 dB
8 500 192 0.4 -8.31 dB
10 500 215 0.4 -7.33 dB
20 500 273 0.5 -5.26 dB
40 500 352 0.7 -3.05 dB
80 500 399 0.8 -1.96 dB
100 500 411 0.8 -1.70 dB
200 500 438 0.9 -1.15 dB
400 500 436 0.9 -1.19 dB
800 500 400 0.8 -1.94 dB
1000 500 391 0.8 -2.14 dB
2000 500 344 0.7 -3.25 dB
4000 500 284 0.6 -4.91 dB
8000 500 203 0.4 -7.83 dB
10000 500 194 0.4 -8.22 dB
20000 500 176 0.4 -9.07 dB
40000 500 146 0.3 -10.69 dB
80000 500 121 0.2 -12.32 dB
100000 500 117 0.2 -12.62 dB

Table 2.2. 343 Hz Frequency Response Data

343 Hz Frequency Response in dB
Figure 2.3. 343 Hz Set Point Circuit Frequency Response in dB

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 500 50 0.1 -20.00 dB
2 500 55 0.1 -19.17 dB
4 500 69 0.1 -17.20 dB
8 500 102 0.2 -13.81 dB
10 500 119 0.2 -12.47 dB
20 500 169 0.3 -9.42 dB
40 500 239 0.5 -6.41 dB
80 500 338 0.7 -3.40 dB
100 500 354 0.7 -3.00 dB
200 500 395 0.8 -2.05 dB
400 500 413 0.8 -1.66 dB
800 500 427 0.9 -1.37 dB
1000 500 424 0.8 -1.43 dB
2000 500 418 0.8 -1.56 dB
4000 500 411 0.8 -1.70 dB
8000 500 402 0.8 -1.89 dB
10000 500 391 0.8 -2.14 dB
20000 500 359 0.7 -2.88 dB
40000 500 268 0.5 -5.42 dB
80000 500 150 0.3 -10.46 dB
100000 500 136 0.3 -11.31 dB

Table 2.3. 1.08 kHz Frequency Response Data

1.08 kHz Frequency Response in dB
Figure 2.4. 1.08 kHz Set Point Circuit Frequency Response in dB

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 500 58 0.1 -18.71 dB
2 500 62 0.1 -18.13 dB
4 500 71 0.1 -16.95 dB
8 500 82 0.2 -15.70 dB
10 500 87 0.2 -15.19 dB
20 500 106 0.2 -13..47 dB
40 500 136 0.3 -11.31 dB
80 500 212 0.4 -7.45 dB
100 500 232 0.5 -6.67 dB
200 500 294 0.6 -4.61 dB
400 500 355 0.7 -2.97 dB
800 500 404 0.8 -1.85 dB
1000 500 414 0.8 -1.64 dB
2000 500 435 0.9 -1.21 dB
4000 500 442 0.9 -1.07 dB
8000 500 405 0.8 -1.83 dB
10000 500 391 0.8 -2.14 dB
20000 500 359 0.7 -2.88 dB
40000 500 252 0.5 -5.95 dB
80000 500 167 0.3 -9.53 dB
100000 500 132 0.3 -11.57 dB

Table 2.4. 3.43 kHz Frequency Response Data

3.43 kHz Frequency Response in dB
Figure 2.5. 3.43 kHz Set Point Circuit Frequency Response in dB

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 500 59 0.1 -18.56 dB
2 500 62 0.1 -18.13 dB
4 500 64 0.1 -17.86 dB
8 500 70 0.1 -17.08 dB
10 500 76 0.2 -16.36 dB
20 500 84 0.2 -15.49 dB
40 500 93 0.2 -14.61 dB
80 500 114 0.2 -12.84 dB
100 500 127 0.3 -11.90 dB
200 500 168 0.3 -9.47 dB
400 500 216 0.4 -7.29 dB
800 500 288 0.6 -4.79 dB
1000 500 328 0.7 -3.66 dB
2000 500 375 0.8 -2.50 dB
4000 500 402 0.8 -1.89 dB
8000 500 420 0.8 -1.51 dB
10000 500 428 0.9 -1.35 dB
20000 500 417 0.8 -1.58 dB
40000 500 400 0.8 -1.94 dB
80000 500 354 0.7 -3.00 dB
100000 500 336 0.7 -3.45 dB

Table 2.5. 10.8 kHz Frequency Response Data

10.8 kHz Frequency Response in dB
Figure 2.6. 10.8 kHz Set Point Circuit Frequency Response in dB

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 500 402 0.8 -1.89 dB
2 500 400 0.8 -1.94 dB
4 500 406 0.8 -1.81 dB
8 500 433 0.9 -1.25 dB
10 500 406 0.8 -1.81 dB
20 500 427 0.9 -1.37 dB
40 500 411 0.8 -1.70 dB
80 500 422 0.8 -1.47 dB
100 500 433 0.9 -1.25 dB
200 500 419 0.8 -1.54 dB
400 500 404 0.8 -1.85 dB
800 500 433 0.9 -1.25 dB
1000 500 434 0.9 -1.23 dB
2000 500 392 0.8 -2.11 dB
4000 500 402 0.8 -1.89 dB
8000 500 408 0.8 -1.77 dB
10000 500 425 0.9 -1.41 dB
20000 500 422 0.8 -1.47 dB
40000 500 408 0.8 -1.77 dB
80000 500 426 0.9 -1.39 dB
100000 500 414 0.8 -1.64 dB

Table 2.6. Overall Circuit Frequency Response Data

Overall Frequency Response in dB
Figure 2.7. Overall Circuit Frequency Response in dB

Power Amplifier

Input Signal:
The signal is set to sine wave with 500 \(mV_{rms}\), 1 kHz, 0 V offset applied at the input.

Output Waveform:

Output Waveform
Figure 3.2. Output Waveform on Oscilloscope

Gain:

\[\textrm{Ratio Gain}=\frac{\textrm{Output RMS}}{\textrm{Input RMS}}=\frac{14.40 V}{500 mV}=\underline{28.8}\] \[\textrm{dB Gain}=20 log_{10} \textrm{Ratio Gain}=20 log_{10} 28.8=\underline{29.19dB}\]

Maximum Average Power Output:

\[P_{average}^{max}=\frac{(V_{rms}^{max})^2}{R_{load}}=\frac{(14.40 V)^2}{8\omega}=\underline{25.92 W}\]

Heat Dissipation:

\[\theta_{sa}=\frac{(T_{jmax}-T_{amb})-P_{dmax}(\theta_{jc}+\theta_{cs})}{P_{dmax}}\]

\(\theta_{sa}=\textrm{Maximum thermal resistance of heat sink}\)
\(T_{jmax}=\textrm{Maximum junction temperature=150℃ (from Datasheet)}\)
\(T_{amb}=\textrm{Ambient temperature=25℃ (Typical Room Temperature)}\)
\(P_{dmax}=\textrm{Maximum power dissipation=25.92W (see above)}\)
\(\theta_{jc}=\textrm{Thermal resistance from junction to case=0.21℃/W (from thermal paste)}\)
\(\theta_{cs}=\textrm{Thermal resistance from case to heat sink=2℃/W (from Datasheet)}\)

So, \(\theta_{sa}=\underline{2.6125℃/W}\)

Heat sink used has thermal resistance of 0.80 ℃/W, which is smaller than the maximum value, as required. During test, chip surface temperature maintained at 51.8℃ at 25.92 W output, within maximum operating temperature range (150℃).

Integrated System

Input Signal:
The signal is set to sine wave with 4 \(mV_{rms}\), 1 kHz, 0 V offset applied at the input.

Output Waveform

Output Waveform at 1 kHz
Figure 4.1. Output Waveform on Oscilloscope (at 1 kHz)

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 4 1.36 340 50.63 dB
2 4 1.44 360 51.13 dB
4 4 2.24 560 54.96 dB
8 4 4.80 1200 61.58 dB
10 4 6.00 1500 63.52 dB
20 4 10.00 2500 67.96 dB
40 4 12.80 3200 70.10 dB
80 4 13.60 3400 70.63 dB
100 4 13.60 3400 70.63 dB
200 4 14.00 3500 70.88 dB
400 4 13.60 3400 70.63 dB
800 4 13.20 3300 70.37 dB
1000 4 13.20 3300 70.37 dB
2000 4 12.00 3000 69.54 dB
4000 4 9.00 2250 67.04 dB
8000 4 4.60 1150 61.21 dB
10000 4 4.08 1020 60.17 dB
20000 4 4.00 1000 60.00 dB
40000 4 3.92 980 59.82 dB
80000 4 2.08 520 54.32 dB
100000 4 1.60 400 52.04 dB

Table 4.1. Frequency Response Data of Output Waveform (20 Hz~20 kHz Filter at Pre-amp)

Output Frequency Response Ratio
Figure 4.2. Frequency Response of Output Waveform (20 Hz~20 kHz Filter at Pre-amp)

Output Frequency Response dB
Figure 4.3. Frequency Response of Output Waveform in dB (20 Hz~20 kHz Filter at Pre-amp)

THD measured at the input to the power stage at 1 kHz = 0.0635%

Frequency Response of Filters at Output Stage:

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 4 1.52 380 51.60 dB
2 4 1.60 400 52.04 dB
4 4 2.60 650 56.26 dB
8 4 5.20 1300 62.28 dB
10 4 6.80 1700 64.61 dB
20 4 11.60 2900 69.25 dB
40 4 14.40 3600 71.13 dB
80 4 14.80 3700 71.36 dB
100 4 15.60 3900 71.82 dB
200 4 16.00 4000 72.04 dB
400 4 16.00 4000 72.04 dB
800 4 15.60 3900 71.82 dB
1000 4 15.60 3900 71.82 dB
2000 4 14.00 3500 70.88 dB
4000 4 10.40 2600 68.30 dB
8000 4 5.40 1350 62.61 dB
10000 4 4.60 1150 61.21 dB
20000 4 4.56 1140 61.14 dB
40000 4 4.40 1100 60.83 dB
80000 4 2.32 580 55.27 dB
100000 4 1.76 440 52.87 dB

Table 4.2. Frequency Response Data (108 Hz Filter)

Output 108 Hz Filter Frequency Response in dB
Figure 4.4. Frequency Response in dB (108 Hz Filter)

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 4 1.52 380 51.60 dB
2 4 1.60 400 52.04 dB
4 4 2.40 600 55.56 dB
8 4 4.80 1200 61.58 dB
10 4 6.20 1550 63.81 dB
20 4 10.80 2700 68.63 dB
40 4 14.00 3500 70.88 dB
80 4 14.80 3700 71.36 dB
100 4 14.80 3700 71.36 dB
200 4 15.60 3900 71.82 dB
400 4 15.60 3900 71.82 dB
800 4 15.60 3900 71.82 dB
1000 4 15.20 3800 71.60 dB
2000 4 13.60 3400 70.63 dB
4000 4 10.00 2500 67.96 dB
8000 4 5.20 1300 62.28 dB
10000 4 4.60 1150 61.21 dB
20000 4 4.60 1150 61.21 dB
40000 4 4.40 1100 60.83 dB
80000 4 2.32 580 55.27 dB
100000 4 1.76 440 52.87 dB

Table 4.3. Frequency Response Data (343 Hz Filter)

Output 343 Hz Filter Frequency Response in dB
Figure 4.5. Frequency Response in dB (343 Hz Filter)

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 4 1.52 380 51.60 dB
2 4 1.60 400 52.04 dB
4 4 2.40 600 55.56 dB
8 4 4.80 1200 61.58 dB
10 4 6.00 1500 63.52 dB
20 4 10.00 2500 67.96 dB
40 4 13.60 3400 70.63 dB
80 4 14.40 3600 71.13 dB
100 4 14.40 3600 71.13 dB
200 4 14.80 3700 71.36 dB
400 4 15.20 3800 71.60 dB
800 4 15.40 3850 71.71 dB
1000 4 15.60 3900 71.82 dB
2000 4 15.40 3850 71.71 dB
4000 4 12.60 3150 69.97 dB
8000 4 7.00 1750 64.86 dB
10000 4 5.20 1300 62.28 dB
20000 4 4.40 1100 60.83 dB
40000 4 4.40 1100 60.83 dB
80000 4 2.40 600 55.56 dB
100000 4 1.80 450 53.06 dB

Table 4.4. Frequency Response Data (1.08 kHz Filter)

Output 1.08 kHz Filter Frequency Response in dB
Figure 4.6. Frequency Response in dB (1.08 kHz Filter)

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 4 1.52 380 51.60 dB
2 4 1.60 400 52.04 dB
4 4 2.40 600 55.56 dB
8 4 4.80 1200 61.58 dB
10 4 6.00 1500 63.52 dB
20 4 10.00 2500 67.96 dB
40 4 12.80 3200 70.10 dB
80 4 14.00 3500 70.88 dB
100 4 14.00 3500 70.88 dB
200 4 15.20 3800 71.60 dB
400 4 15.60 3900 71.82 dB
800 4 15.60 3900 71.82 dB
1000 4 15.60 3900 71.82 dB
2000 4 14.00 3500 70.88 dB
4000 4 10.00 2500 67.96 dB
8000 4 5.20 1300 62.28 dB
10000 4 4.48 1120 60.98 dB
20000 4 5.04 1260 62.01 dB
40000 4 4.24 1060 60.51 dB
80000 4 2.28 570 55.12 dB
100000 4 1.76 440 52.87 dB

Table 4.5. Frequency Response Data (3.43 kHz Filter)

Output 3.43 kHz Filter Frequency Response in dB
Figure 4.7. Frequency Response in dB (3.43 kHz Filter)

Frequency / Hz Input RMS / mV Output RMS / mV Ratio Gain dB Gain
1 4 1.52 380 51.60 dB
2 4 1.60 400 52.04 dB
4 4 2.40 600 55.56 dB
8 4 4.80 1200 61.58 dB
10 4 6.00 1500 63.52 dB
20 4 10.00 2500 67.96 dB
40 4 12.80 3200 70.10 dB
80 4 13.20 3300 70.37 dB
100 4 13.60 3400 70.63 dB
200 4 14.00 3500 70.88 dB
400 4 14.00 3500 70.88 dB
800 4 14.40 3600 71.13 dB
1000 4 14.80 3700 71.36 dB
2000 4 16.40 4100 72.26 dB
4000 4 18.40 4600 73.26 dB
8000 4 20.00 5000 73.98 dB
10000 4 20.40 5100 74.15 dB
20000 4 13.60 3400 70.63 dB
40000 4 4.40 1100 60.83 dB
80000 4 2.20 550 54.81 dB
100000 4 1.80 450 53.06 dB

Table 4.6. Frequency Response Data (10.8 kHz Filter)

Output 10.8 kHz Filter Frequency Response in dB
Figure 4.8. Frequency Response in dB (10.8 kHz Filter)


Conclusion

Mic Preamplifier

The preamp circuit meets all the design specification and requirements.

The frequency response shows a clear band-pass filter pattern with cutoff frequency close to 20Hz and 20kHz, meeting the design purpose. The power of signal only loses 0.35dB during the band-pass filtering process, which is acceptable.

The SNR calculation shows a result of 45.68 dB, which is significant, but it’s still acceptable considering the prototyping method of breadboard and bridging wires.

5 Band Equalizer

The 5-band equalizer circuit meets all the design specification and requirements.

The frequency response shows five results with peak frequency close to the preset values, meeting the design purpose. The frequency response for overall circuit shows a nearly flat pattern in large scale, and has an average gain of -1.61 dB, which is within the range stated in the IC datasheet, -3.8 dB ~ 2.2 dB.

The total harmonic distortion of the chip is only 0.02%, which is pretty low and decent in audio processing.

Power Amplifier

The power amplifier circuit meets all the design specification and requirements.

The output power reach 25.92 W at maximum, meeting the design purpose of over 20 W. The gain is 28.8 in test, with 7.09% error from the designed gain of 31, which is acceptable.

Temperature dissipation meets all the speculations through calculation, and in real test has a good performance of maintaining 51.8℃ surface temperature, which is significantly good.

Integrated System

The integrated system meets all the design purpose.

The frequency response at output stage shows a fine result of the 20 Hz~20 kHz band-pass filter at pre-amp stage, which turns out to do a great job on eliminating wireless communication noises out of human hearing range.

The frequency response for the five equalizer filters at output stage shows effective equalizing functions and doesn’t influence the overall signal power excessively.

The total harmonic distortion of the circuit before the power stage is only 0.0635% at 1 kHz input, which is significantly good. Even added the little distortion caused by the single-chip-solution class AB amplifier, the output audio signal is good enough with negligible noise.