Projects per year
Audio reproduction systems contains two key components, the amplifier and the loudspeaker. In the last 20 – 30 years the technology of audio amplifiers have performed a fundamental shift of paradigm. Class D audio amplifiers have replaced the linear amplifiers, suffering from the well-known issues of high volume, weight, and cost. High efficient class D amplifiers are now widely available offering power densities, that their linear counterparts can not match. Unlike the technology of audio amplifiers, the loudspeaker is still based on the traditional electrodynamic transducer invented by C.W. Rice and E.W. Kellog in 1925 . The poor efficiency of the electrodynamic transducer remains a key issue, and a significant limit of the efficiency of the complete audio reproduction systems. Also the geometric limits of the electrodynamic transducer imposes significant limits on the design of loudspeakers. The challenge of designing a flat loudspeaker based on the electrodynamic transducer is still not fulfilled. Alternatives to the electrodynamic transducer based loudspeaker is the piezoelectric, horn, electrostatic and distributed-mode loudspeaker. The directivity of the electrostatic loudspeaker combined with the low level of acoustical output power and complex amplifier requirements, have limited the commercial success of the technology. Horn or compression drivers are typically favoured, when high acoustic output power is required, this is however at the expense of significant distortion combined with a large volume of the loudspeaker enclosure. Piezoelectric loudspeakers suffers from the poor power handling capability of the piezoelectric ceramic. However a niche is found in the market of hydrophones, because of the excellent impedance matching between the piezoelectric transducer and water. Distributed-mode loudspeakers represent a very interesting attempt for designing flat loudspeakers. The poor bass response combined with the diffuse and uncorrelated acoustic output, remains a challenge [2, 3]. The work presented focuses on the development of an amplifier for a special type of transducer, the DEAP (Dielectric ElectroActive Polymer) one. DEAP based loudspeakers work on the principle of the electrostatic forces, and possess some of the ii same characteristics as the electrostatic loudspeaker. However, the DEAP transducer is constructed by printing compliant, corrugated electrodes on a silicone film. As a consequence a capacitive transducer emerges, which can be shaped into the loudspeaker membrane itself, rolled up into a transducer driving a membrane or being part of an active suspension system for the membrane. In order to document the full potential of the DEAP transducer, suitable amplifiers must be developed. The frequency response and linearity of these amplifier is essential, as the application considered is that of audio. Also the efficiency of the amplifier is a key concern. An introduction to the project is given in chapter 1, followed by a state-of-the-art study in chapter 2. Due to the similarities between the electrostatic loudspeaker and the DEAP transducer, the state-of-the-art has a special focus on amplifiers for electrostatic loudspeakers. Amplifiers for other type of capacitive transducers like piezoelectric ones are also considered. Finally the current state-of-the-art for class D audio amplifiers driving the electrodynamic transducer is presented. Chapter 3 gives an introduction to the DEAP transducer as a load in loudspeaker systems. The main purpose being to established the frequency response of the DEAP input impedance, but also investigate the large signal implications of driving the non-linear transducer of the DEAP. 2-level modulated high voltage amplifiers driving the capacitive load of the DEAP transducer are addressed in chapter 4. An amplifier with fourth order output filter and full-state self-oscillating hysteresis based control loop is proposed. The control loop ensures high open loop gain and active damping. Active damping is a key feature in order to achieve high amplifier efficiency. In order to further increase the output voltage or reduce the semiconductor voltage stress, multilevel inverters as amplifiers for class D audio amplifiers was introduced. The flying capacitor three-level modulated inverter is analysed, implemented and tested. A control scheme is proposed allowing for the balancing of the flying capacitor, while ensuring active damping. This subject is covered in chapter 5. It is concluded, that class D audio amplifiers for high voltage capacitive transducers can be constructed with THD+N below 0.1 % and peak efficiency above 80 %. However the complexity of the amplifier combined with the current high cost of components, makes the technology of DEAP based loudspeaker unfeasible. Suggestions to future work in the pursuit of successful commercialisation of the DEAP technology for audio applications is given in the final chapter.
|Publisher||Technical University of Denmark, Department of Electrical Engineering|
|Number of pages||169|
|Publication status||Published - 2014|