Intermodulation Distortions of Bang–Bang Control Class D Amplifiers

Guo, Linfei; Ge, Ting; Chang, Joseph S.

doi:10.1109/tpel.2014.2305432

Cited by 18 publications

(22 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fig. 7 shows the variation of the calculated switching frequency (F SW ) versus the duty cycle (D) for several τ d cases as expressed in (10). It can be seen that F SW is a parabolic function of D, and the delay τ d would impose a limit in the maximum achievable F SW .…”

Section: Self-oscillating Modulator Considerationsmentioning

confidence: 99%

“…The degradation in THD+N in filter (V) appears to be caused by the ferrite bead material due to its magnetic history curve (B-H curve) non-linear behavior, and a non-constant permeability (µ m ) that changes with the magnitude of the magnetic field and operating frequency [44]. The THD+N for the output filter (III) with a signal at 6.67 Power-supply intermodulation distortion (PS-IMD) provides a metric to evaluate the effect of the amplifier's power-supply noise when the audio signal is also present [9], [10], [12]. A 1 kHz sine wave with 0.5 V P P was used as the input of the audio amplifier together with 0.2 V P P at 217 Hz signal at the amplifier's high-voltage supply V CC .…”

Section: F Output Filter Considerationsmentioning

confidence: 99%

“…To achieve outstanding audio performance in a CDA, closed-loop architectures are typically used where the negative feedback mechanism helps to correct errors in the amplification process. The closed-loop CDA has been analyzed for intermodulation distortion (IMD) in time domain [9] and in frequency domain [10]. Moreover, the carrier distortion and its effect on the system has been analyzed in [11], and the effect of power-supply noise was analyzed in [12].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A High-Efficiency Self-Oscillating Class-D Amplifier for Piezoelectric Speakers

Colli-Menchi

Sánchez-Sinencio

2015

IEEE Trans. Power Electron.

View full text Add to dashboard Cite

The design tradeoffs of the class-D amplifier (CDA) for driving piezoelectric (PZ) speakers are presented, including efficiency, linearity, and electromagnetic interference. An implementation is proposed to achieve high efficiency in the CDA architecture for PZ speakers to extend battery life in mobile devices. A self-oscillating closed-loop architecture is used to obviate the need for a carrier signal generator to achieve low power consumption. The use of stacked-cascode CMOS transistors at the H-bridge output stage provides low input capacitance to allow high switching frequency to improve linearity with high efficiency. Moreover, the CDA monolithic implementation achieves 18 V P P output voltage swing in a low voltage CMOS technology without requiring expensive high-voltage semiconductor devices. The prototype experimental results achieved a minimum THD+N of 0.025%, and a maximum efficiency of 96%. Compared to available CDA for PZ speakers, the proposed CDA achieved higher linearity, lower power consumption, and higher efficiency. a larger silicon area; thus, increasing the cost of the amplifier. Commercial CDA architectures for PZ speakers provide high-voltage outputs using these devices, but the distortion and power consumption is still large [27]- [30].Other switching output stages have been proposed to drive high-voltage capacitive actuators for different applications [31], [32]. Nonetheless, the primary objective of these applications is to deliver the maximum amount of energy at the actuator's resonant point, making them not suitable for audio applications.This work discusses the design tradeoffs of the CDA architecture for driving PZ speakers, especially when low power consumption and high efficiency are desired. An example implementation is proposed to achieve high efficiency and high linearity in the CDA architecture for PZ speakers to extend battery life in mobile devices. The self-oscillating closed-loop architecture is used to obviate the need of a carrier signal generator to achieve high linearity with low power consumption. Moreover, the CDA monolithic implementation is able to provide an 18 V pp output voltage swing in an 1.8 V core-voltage twin-well 11-16 Ω-cm p-type substrate CMOS technology without requiring expensive special high-voltage semiconductor devices. The use of stacked-cascode CMOS transistors at the H-bridge output stage provides low input capacitance to allow high switching frequency to improve linearity without sacrificing the high efficiency. This paper is organized as follows: Section II reviews the PZ speaker details and design considerations for the CDA. Section III discuses the CDA architecture for PZ speakers and its tradeoffs. Section IV presents the measurement results of the monolithic implementation prototype, and Section V concludes the paper. II. PIEZOELECTRIC SPEAKERS A. Structure and operationThe physical structure of a typical PZ speaker is shown in Fig. 2 where a PZ element is attached to a film encased between a front panel and rear panel. The PZ element deflect...

show abstract

Section: Self-oscillating Modulator Considerationsmentioning

confidence: 99%

Section: F Output Filter Considerationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A High-Efficiency Self-Oscillating Class-D Amplifier for Piezoelectric Speakers

Colli-Menchi

Sánchez-Sinencio

2015

IEEE Trans. Power Electron.

View full text Add to dashboard Cite

show abstract

“…A significant achievement of the present work is that such expressions are given for a general audio input. While results are illustrated for the important test case of a sinusoidal input, from which the THD is calculated, the general expression provides a valuable tool -not previously available -for engineers who wish to calculate measures of distortion beyond just the THD, such as the intermodulation distortion (IMD) (Cox et al, 2011Guo et al, 2014;Metzler, 1993;Yu et al, 2012) that occurs when the input contains two or more frequencies.…”

Section: Introductionmentioning

confidence: 99%

Analysis of a hysteresis-controlled self-oscillating class-D amplifier

Cox¹,

Yu²,

Goh³

et al. 2016

IMA J Appl Math

View full text Add to dashboard Cite

This paper gives the first systematic perturbation analysis of the audio distortion and mean switching period for a self-oscillating class-D amplifier. Explicit expressions are given for all the principal components of audio distortion, for a general audio input signal; the specific example of a sinusoidal input is also discussed in some detail, yielding an explicit closed-form expression for the total harmonic distortion (THD). A class-D amplifier works by converting a low-frequency audio input signal to a high-frequency train of rectangular pulses, whose widths are slowly modulated according to the audio signal. The audiofrequency components of the pulse-train are designed to agree with those of the audio signal. In many varieties of class-D amplifier, the pulse-train is generated using a carrier wave of fixed frequency, well above the audio range. In other varieties, as here, there is no such fixed-frequency clock, and the local frequency of the pulse-train varies in response to the audio input. Such self-oscillating designs pose a particular challenge for comprehensive mathematical modelling; we show that in order to properly account for the local frequency variations, a warped-time transformation is necessary. The systematic nature of our calculation means it can potentially be applied to a range of other self-oscillating topologies. Our results for a general input allow ready calculation of distortion diagnostics such as the intermodulation distortion (IMD), which prior analyses, based on sinusoidal input, cannot provide.

show abstract

“…At this juncture, the design-art for audio CDAs is mature -to the point where CDAs featuring high power-efficiency and high fidelity attributes are somewhat routine [5][6][7][8][9][10][11][12][13][14]; the 'high-fidelity' are often defined in terms of high power supply rejection ratio (PSRR) [15][16][17][18][19][20][21], low power supply induced intermodulation distortion (PS-IMD) [20][21], low total harmonic distortion plus noise (THD+N) [22][23][24], low intermodulation distortion (IMD) [25][26][27] and low electromagnetic interference (EMI) [28][29][30][31]; see Chapter 2 later. Nevertheless, in a smartphone, the operating condition for the CDA in the context of noise is challenging.…”

Section: Background and Motivationsmentioning

confidence: 99%

Analysis and design of switching amplifiers for smartphones

He¹

View full text Add to dashboard Cite

The past years as a Ph.D. student have been a special journey for me with many fond memories. Without the kind assistance of many people, it would have been very difficult for me to achieve this work. First of all, I would like to express my gratitude to my advisor, Prof. Joseph Chang. Prof. Chang guided me throughout my entire Ph.D. program with his outstanding academic wisdom and valuable suggestions. His support throughout these years helped me to move forward. It was one of my most enjoyable experiences learning from his profound knowledge and expertise. I would like to thank my co-advisor, Dr. Yu Rongshan. He offered valuable guidance as well as kind support throughout this Ph.D. program. I am especially grateful to Dr. Ge Tong for discussions pertaining to the technical aspects in this research program.

show abstract

Intermodulation Distortions of Bang–Bang Control Class D Amplifiers

Cited by 18 publications

References 13 publications

A High-Efficiency Self-Oscillating Class-D Amplifier for Piezoelectric Speakers

A High-Efficiency Self-Oscillating Class-D Amplifier for Piezoelectric Speakers

Analysis of a hysteresis-controlled self-oscillating class-D amplifier

Analysis and design of switching amplifiers for smartphones

Contact Info

Product

Resources

About