Gael Pillonnet

Optimal state-space controller for power switching converter

This paper describes the state-space controller for the buck converter working in Continuous Conduction Mode (CCM). The converter operates at a switching frequency of 500 KHz. Aiming to study the control of this kind of converter, some classical control techniques and the control in the state space are discussed. As an example, we present the design procedure for a dc-dc buck converter with an analog controller in s-domain, and a linear quadratic regulator (LQR). The controller is robust to the load variation and the change in supply and reference voltage. The performance of the controllers is compared and conclusions are drawn. Matlab/Simulink simulated results illustrate that the control system has good static and dynamic characteristics.

Comparison between voltage and current driving methods of a micro-speaker☆

Abstract Since its inception, most audio amplifiers control the loudspeaker in voltage. However previous studies highlighted the importance of the loudspeaker control in current. These studies have been done only with large size loudspeakers (bass or midrange loudspeakers) and this is certainly not transposable for the type of loudspeaker in interest i.e. micro-speaker. First of all, this paper describes a model of loudspeaker (voltage driven and also current driven) represented by a comprehensive set of data based on a minimal number of measurements. Simulation results based on these models are presented using single frequency signals such as multi-frequencies signals to compare the two driven methods. At this level of modelling, simulation results show that, contrary to the woofer applications, current driving of micro-speaker does not affect significantly in terms of harmonic distortions, intermodulation distortions and transient behavior.

A high PSRR Class-D audio amplifier IC based on a self-adjusting voltage reference

In a wide range of applications, audio amplifiers require a large Power Supply Rejection Ratio (PSRR) that the current Class-D architecture cannot reach. This paper proposes a self-adjusting internal voltage reference scheme that sets the bias voltages of the amplifier without losing on output dynamics. This solution relaxes the constraints on gain and feedback resistors matching that were previously the limiting factor for the PSRR. Theory of operation, design and IC evaluation in a Class-D amplifier in CMOS 0.25µm will be shown in this paper. The use of this voltage reference increased the amplifiers PSRR by 15dB, with only a 140µA increase in current consumption.

N-Conductor Passive Circuit Modeling for Power Converter Current Prediction and EMI Aspect

This paper proposes a method to study the impact of passive circuits such as filters, loads, and board layout on conducted electromagnetic emissions. The method is dedicated to power converters with N-active conductors and a ground conductor, followed by passive circuits. A sophisticated use of impedance matrices allows high-frequency current prediction. The power system is divided into functional circuits called blocks. Each passive block is modeled by an impedance matrix. A fast matrix calculation permits association of those matrices to obtain a compact model represented by an impedance matrix. Knowing the converter output voltages and the resulting impedance matrix, currents are calculated to study the influence of each passive block on high-frequency current spectrum, and so the electromagnetic interference. This will help to filter and board layout design for a given load. After presenting the general use of the method, the validation is performed on a differential Class-D audio amplifier used in cell phone applications. The practical application on a two-active conductor system validates the method up to 110 MHz.

Conducted EMI prediction for integrated class D audio amplifier

The aim of this paper is to predict the delivered currents at the output of an integrated audio switching amplifier for EMI prediction. Impedance matrices are used to model the different passive parts of the system. Hereafter, all the matrices are associated in a single one, where the resulting matrix and the output voltages in open circuit are used to predict the output currents spectra directly in the frequency domain. This method can be used by system designers and system integrators in order to study their systems EM emissions before assembling the different parts of the system. The experimental application of this method gives good accuracy up to 10 MHz (twenty times the switching frequency).

Loss analysis of flyback in discontinuous conduction mode for sub-mW harvesting systems

This energy harvesting solution permits the autonomous supply of low power sensor nodes, without a chemical battery, allowing their extensive use in various environments. The electrical interface between the harvester and the sensor is crucial in order to maximize the harvested energy and boost the voltage to a minimum value required by the sensor. To achieve input impedance and voltage gain independently, this paper presents a flyback converter in discontinuous mode. Using the proposed flyback model validated experimentally, we have studied the impact of each loss source in order to give some trade-off for designing an efficient sub-mW harvesting interface. We underline the effect of the transformer losses due to the magnetic hysteresis as well as the driving loss impact. Following this method, the flyback prototype achieves 71% power efficiency when harvesting from a microbial fuel cell delivering 90μW.

Predicting the Impact of Magnetic Components Used for EMI Suppression on the Base-Band of a Power Amplifier

Class-D audio amplifiers are switching circuits that produce serious Electromagnetic (EM) emissions and disturb the surrounding electronics. In order to reduce these emissions, electromagnetic compatibility (EMC) filters with ferrite beads are used. However, ferrite beads contain magnetic materials that have a nonlinear behavior. Thus, they have an unfavorable impact on the system audio quality. The common ferrite bead models do not take into account nonlinear phenomena. Thus, to predict the impact on the signal quality, this paper models the ferrite bead using the Jiles-Atherton magnetic material theory. The presented model provides the designers with a tool to quantify the effect of EMC filters on the total harmonic distortion (THD) of audio amplifiers. The simulated and measured results show that the tested ferrite bead have a negative effect on the audio signal for a wide range of amplitudes and can increase the THD up to 37 dB. Finally, this paper highlights the impact of the magnetic material type on the audio distortion by simulating the same component with different types of materials.

A class D headphone amplifier with DC coupled outputs and 1.2mA quiescent current

This paper shows the design of a headphone amplifier using Pulse Width Modulation (PWM) to improve its power efficiency compared to existing Class AB and Class G structures. A numerical optimization of the switching frequency and output transistors size, combined with the use of a soft switching scheme, brought the current consumption down to 1.2mA at idle for the complete stereo amplifier. The outputs are DC coupled, with no common mode voltage, thanks to the use of a Negative Charge Pump (NCP). Theory of operation, optimization, and the design in a 0.25μm CMOS process will be explained.

Improved Class-K amplifier for headset applications

This paper proposes an improved Class-K audio amplifier as an alternative solution to Class-AB and Class-G audio amplifiers for Headset applications. The Class-K architecture consists in a linear and a switching amplifier connected in parallel to the same speaker. The advantage of this solution is to associate the power efficiency of the switching amplifier with the linearity of a linear amplifier. However, the Class-K may suffer from static consumption. Three methods are presented in order to compensate the topologys disadvantages: a ternary modulation, a two-stage linear amplifier and a low idle switching amplifier. In this paper, we present the Class-K operating principle, a topological comparison and solutions to improve high static consumption. The simulation results prove the validity of these solutions, achieving less than 2mA of static current.

A synchronized self oscillating Class-D amplifier for mobile application

An integrated Class-D amplifier using a synchronized self-oscillating modulator based on a hysteresis control is presented. This new feedback topology can achieve a higher gain bandwidth product to increase linearity and power supply rejection than the fixed switching frequency solutions. The switching frequency spread can also reduce the external EMI filtering constraints. To reduce the idle consumption and radiated emissions at zero output level, a synchronization technique is presented to provide truly differential outputs. The proposed amplifier, realized in CMOS 130nm process, drives up to 1W load with 0.02% THD+N, 90% efficiency with power supply range from 2.3 to 5V. The synchronization technique reduces shows up to 75% of idle consumption.