低功率低失真無濾波級之D類功率放大器

Abstract

放大器對一個音響系統扮演著很重要的角色。在目前音響系統中市場仍以AB類放大器為主要設計。若以功率效益方面來比較，D類功率放大器有較優越的輸出功率表現。主要是因為D類功率放大器是以切換的方式來完全導通或是完全關閉電晶體。因此D類功率放大器可以擁有AB類放大器較少的功率消耗。 本論文重心著重於發展一種高功率效益D類功率放大器的調變方式。此電路有別於傳統全橋式D類功率放大器，它改善了傳統D類功率放大器訊號失真過大的缺點，同時卻不因此使得功率效益變小；改善方法是藉由加入回授系統；這是一種降低雜訊的技巧；並且改變脈衝寬度調變的方式來實現電路。此外，所設計的電路還有一項優點，就是不需要D類功率放大器的輸出濾波級，此技術使得電路的面積變小，功率效益提高。本電路是使用台積電製程0.35μm 2P4M（混合訊號）來模擬與設計。整個系統架構包括“兩個誤差放大器＂，“兩個比較器＂，“一個三角波產生器＂，“緩衝器＂，“閘極驅動級＂，“H－橋＂等電路方塊。 D類功率放大器模擬結果，輸入頻率2KHz，振幅0.7V的正弦波，操作在250KHz，截止頻率20KHz。輸出結果在功率效益得到85％，總諧波失真值為0.12％，整體消耗功率在12.57mW。 經由各種調變設計結果與模擬，可以明白比較出四種調變方法的優缺點。若欲使THD＜1％，則最好選擇輸入訊號振幅為載波訊號振幅的0.7倍以內會有較佳結果。實際上，一般D類功率放大器功率效益大約是在85％~95％。若要在提高D類功率放大器的功率效益則必須用較大的晶片面積來換取。

Amplifiers play an important role in the audio system. Class-AB amplifiers seem to be the dominative design in recent markets. Compared by power efficiency, Class-D power amplifiers have better performance than Class-AB amplifiers. It is based on the fact that Class-D power amplifiers use the switching operation to fully turn on or off the transistors. Class-D power amplifiers have less power dissipation than Class-AB amplifiers. The focus on this thesis is to develop the modulation of Class-D amplifiers with high power efficiency. This circuit is different from the traditional H-bridge Class-D power amplifier. It has improved the shortcoming of the traditional Class-D power amplifier with too big signal distortion, but not make the power efficiency diminished at the same time. The addition of the feedback system can improve the shortcoming； this is a technology that can reduce noise. And the circuit is realized by changing the way of the Pulse Width Modulation（PWM） method. In addition, this circuit has another advantage that it eliminates the Class-D output filter. This technology can reduce the area of the circuit and can improve power efficiency at the same time. This circuit was fabricated by TSMC 0.35mμ 2P4M（Mixed－Signal）process. It includes two error amplifiers, two comparators, triangular wave generator, buffer, gate-driver, and H-bridge. The Class-D power amplifier has the input sinusoidal-wave frequency of 2 KHz, the amplitude of 0.7 V. It is operated at 250 KHz with the cutoff frequency of 20 KHz. The result shows the power efficiency of 85%, the Total Harmonic Distortion is 0.12%, and the power dissipation is about 12.57 mW. All the modulation simulations are shown in this thesis, and we can obviously compare their advantages and disadvantages. If THD＜1％ is desired, then it is better to choose the amplitude of the input signal to be within 0.7 times of the carrier wave. In fact, the power efficiency of Class-D power amplifier is about 85％ to 95％ generally. If we still would like to increase the power efficiency, the tradeoff is between the power efficiency and the chip area.