Design of audio power amplifier based on NBDD pulse width modulation

Digital audio power amplification technology adopts a new amplification system. The power amplifier tube works in a Class D switching state. Compared with traditional analog power amplifiers, it has the advantages of small size, high power, seamless integration with digital sound sources, effective reduction of transmission interference between signals, and realization of high fidelity. It has broad development prospects.

This paper proposes an optimized design scheme for a high-efficiency digital power amplifier. The double-sideband three-level natural sampling method (NBDD) pulse width modulation technology is introduced into the pulse width modulation design of the digital power amplifier, which reduces the design order of the low-pass filter and improves the signal-to-noise ratio. By introducing the Dead Time technology into the design of the switching amplifier, the series loss and drain-source capacitance loss of the switching amplifier are reduced.

1 Optimization scheme implementation principle

This solution uses two independent channels, which can complete the digital processing and power amplification of the signal separately and simultaneously, and can be bridged into one channel for digital processing and power amplification of the signal. Each channel works in half-bridge working mode and can be bridged into full-bridge working mode. Its implementation principle is shown in Figure 1.

Figure 1 Schematic diagram of a high-efficiency digital power amplifier

The input analog audio signal is first amplified by the isolation amplifier and low-pass filtered at the same time. The low-pass filter uses a second-order Butterworth low-pass filter with a cutoff frequency of 37 kHz and a 3 dB bandwidth of 22 kHz. The filtered signal and the feedback audio signal are sent to the error amplifier for error amplification, and the amplified error audio signal is output. The amplified error signal and the carrier signal are sent to the pulse width modulator for NBDD modulation to generate a PWM signal. The carrier signal is a high-linearity analog triangle wave signal generated by a triangle wave generator with an adjustable frequency of 230~280 kHz. The PWM signal is inserted into the Dead Time and sent to the driver combined with the floating power supply and bootstrap for pre-amplification. The amplified PWM signal drives the half-bridge switching amplifier composed of field effect transistors for power amplification and outputs a power PWM signal. The PWM signal amplified by the switching amplifier is sampled and sent to the error amplifier as a feedback signal.

The power PWM signal is sent to a low-pass filter to restore the analog audio signal.

When a single-channel output needs to be bridged, it is only necessary to input equal-amplitude and anti-phase audio signals to the input ends of the two half-bridges and connect the load to the output ends of the two half-bridges.

In order to increase the reliability of the module, the design also takes into account the damage to the module caused by various misoperations, and provides a fault indication function to help the whole machine find problems in a timely and accurate manner, making it easier to repair the module.

2 Implementation of NBDD modulation technology

The specific implementation of NBDD modulation technology is shown in Figure 2.

Figure 2 NBDD modulation technology implementation block diagram

The input analog audio signal is first amplified by the isolation amplifier, and then sent to the error amplifier together with the feedback audio signal, and the amplified error audio signal is output. The amplified error signal and the carrier signal are sent to the pulse width modulator for NBDD modulation. The carrier signal is a high-linearity analog triangle wave signal generated by the triangle wave generator with a frequency of 230~280 kHz.

The focus here is to achieve a high-linearity triangle wave generator and a high-speed comparator. The nonlinearity of the triangle wave will directly affect the linearity of the PWM modulator and the distortion of the whole machine; in order to restore the audio well, the PWM switching frequency cannot be lower than 200 kHz, so a high-speed comparator is required. The modulation method not only affects the performance indicators within the audio band, but also has a decisive influence on the high-frequency radiation performance (EMC) of the amplifier system. Therefore, from the audio input to the pulse width encoding completion link, the audio amplifier and error amplifier used should have high input impedance, low operating current, wide gain bandwidth, fast rise speed, good common mode rejection ratio, low drift voltage and other technical indicators; the comparator should have the characteristics of fast response speed, low power consumption, and small input offset voltage.

3 Optimal Design of Switching Amplifiers with Dead Time

The main feature of a switching amplifier is its high efficiency, so its optimized design should mainly focus on further reducing various types of losses and truly reflecting its high efficiency characteristics.

Based on the principle of series conduction loss, we can find a way to adjust the gate drive voltage. After the upper tube is completely turned off, the lower tube will start to conduct, and after the lower tube is completely turned off, the upper tube will start to conduct. In this way, the series conduction loss can be reduced, and the junction capacitance Cds loss can be reduced at the same time. This time added between the two drive signals according to the principle of delayed conduction and normal cutoff to solve the series conduction loss is called Dead Time. The principle is shown in Figure 3. The figure analyzes two N-channel field effect tubes working on one switch arm.

Figure 3 Signal comparison before and after the introduction of Deed Time

4. Tests of various indicators

The indicator test mainly uses the internationally used audio tester Audio Precision System One. Audio Precision System One is manufactured by the world's largest audio test instrument manufacturer, American Audio Precision Corporation.

The power is sent to the power socket of the tested sample through the ammeter; a voltmeter is connected in parallel between the positive and negative terminals of the power output. The voltmeter and ammeter are used to test the voltage and current of the power output respectively, so that the power output power can be calculated. The audio input terminal of the tested sample is connected to the audio output terminal of the audio tester, and the power audio output terminal is connected to the audio tester and the standard power resistor respectively. The power signal output by the tested sample is sent to the standard load and sent to the audio tester for analysis and testing. The computer controls the selection of the output signal frequency, amplitude and other characteristics of the audio tester Audio Precision Sy stem One, and selects the indicators to be tested, and displays the test results on the computer.

4.1 Functional index test

Connect the module as normal. If there is no special requirement, the audio input frequency is a 1 kHz sine wave, and the power supply voltage is ±120 V. Test the 8th pin of the XSZ socket. If it is high level (+ SV), it means the module is in protection state, and there is no signal output from the audio output pin; if it is low level (OV), it means the module is in normal working state, and there is signal output from the audio output pin.

The test items and test conditions are:

(1) Mute control: When a mute signal is input, there is no signal output from the audio output pin, the 8th pin of XSZ is high, the module is in protection state, and responds to external mute control;

(2) Power supply overvoltage protection: Increase the +120 V power supply to +128 V, keep the negative power supply unchanged, and the module enters the protection state; reduce the -120 V power supply to -128 V, keep the positive power supply unchanged, and the module enters the protection state;

(3) Power supply undervoltage protection: When the +120 V power supply is reduced to +100 V, the negative power supply remains unchanged, and the module enters the protection state; when the -120 V power supply is increased to -100 V, the positive power supply remains unchanged, and the module enters the protection state;

(4) Power reverse connection protection: When the power supply is connected in reverse, the module will enter the protection state without damage;

(5) Power overcurrent protection: The output standard load is changed to 2Ω, and the input audio signal amplitude is increased. When the output power exceeds 2800W, there is no signal output from the audio output pin, the 8th pin of XS2 is high, and the module is in protection state;

(6) High temperature protection: Use a high temperature oven to heat the module. When the internal temperature of the module reaches +80℃, there is no signal output from the audio output pin, the 8th pin of XS2 is high, and the module is in protection state;

(7) Output to ground protection: Short the module output audio pin to the ground, the audio output pin has no signal output, the 8th pin of XS2 is high, and the module is in protection state;

(8) Output short circuit protection: short-circuit the module output audio pins, no signal is output from the audio output pins, pin 8 of XS2 is high, and the module is in protection state;

(9) Protection indication: When the module enters any protection state, pin 8 of XS2 is high and the module is in protection state.

From the above test results, it can be seen that the digital power amplifier in this paper can meet the requirements of stable operation in many aspects such as mute control, power supply overvoltage protection and power supply undervoltage protection.

4.2 Technical indicator test

The audio input frequency is a 1 kHz sine wave, and the power supply voltage is 120 V. The settings of the audio tester Audio Precision System One are selected by computer control, and test items are selected according to different indicators. The test results are shown in Table 1 and Figures 4 to 6.

Figure 4 1 000 W/ 4 prototype frequency response index test results

Figure 5 1000 W/4 prototype low-level noise index test results

Figure 6 1 000 W/ 4 prototype distortion index test results

4.3 Test results analysis

The test indicators are compared with those of traditional analog amplifiers and first-class foreign digital amplifier manufacturers, and the results are shown in Table 2. From the comparison results, it can be seen that the performance indicators of the prototype are basically consistent with those of professional amplifiers from internationally renowned manufacturers.

The optimized design of the high-efficiency digital power amplifier has been verified to be reasonable and feasible through the research and development of prototypes, and has achieved high-efficiency and high-index amplification of audio signals. It has achieved relatively ideal results in the development of high-power fields. The NBDD pulse width modulation method is adopted to achieve high-quality pulse width modulation, perfectly reproduce the pulse width modulation waveform, and have a high distortion index. Various protection measures taken to improve the reliability of the system have achieved the expected results and improved the reliability of the system.

5 Conclusion

The NBDD modulation method is used to perform pulse width modulation sampling on the audio signal. After adding Dead Time, the pulse width modulation signal is amplified by a driving method combining floating power supply and bootstrap. The amplified pulse width modulation signal drives the switching amplifier in the half-bridge working mode for power amplification, realizing the optimized design of high-efficiency digital audio power amplifier. Various optimization designs adopted in the scheme design have achieved certain results, and the analysis and calculation methods are reasonable and feasible.