Home | Technical Program | Exhibition | Visitors | Students | Press
Last Updated: 20060404, meiP26 - Posters: Loudspeakers and Sound Reinforcement
Tuesday, May 23, 09:00 — 10:30
P26-1 Methods to Improve the Horizontal Pattern of a Line Array Module in the Midrange—Nils Benjamin Schröder, Tobias Schwalbe, Robert Mores, Hamburg University of Applied Sciences - Hamburg, Germany
This paper reviews methods for modeling the vertical directivity of the frequency range from 200 Hz to 1 kHz in line array configurations. It describes the advantages and disadvantages of the following concepts: the horn, the “V-Alignment,“ the flat alignment, and the partial coverage of the loudspeakers. We will shed light on the interrelationship between the angle of two cone loudspeakers and the resulting directivity. Symmetrical and asymmetrical configurations of mid-range drivers and horns are compared. We will outline a procedure to combine these solutions for superior results. One main result will be the desired match of the midsection’s directivity with the directivity of the hf-waveguide section. A concept for building systems with variable directivity over the whole frequency range will be drafted.
[Poster Presentation Associated with Paper Presentation P19-2]
Convention Paper 6776 (Purchase now)
P26-2 The Performance and Restrictions of High Frequency Waveguides in Line Arrays—Nils Benjamin Schröeder, Tobias Schwalbe, Robert Mores, Hamburg University of Applied Sciences - Hamburg, Germany
It is necessary to form a plane coherent wavefront in the hf-section of line arrays. Several different concepts have been applied to reach this goal. We discuss these existing solutions. The different ideas on how to create a cylindrical wavefront will be explained and evaluated. Especially those waveguides that have their weak point in the theoretical design will be criticized. An explanation on how we developed a new waveguide will be given. Finally, we want to give some ideas on how the next generation of waveguides could be designed.
[Poster Presentation Associated with Paper Presentation P19-3]
Convention Paper 6777 (Purchase now)
P26-3 Efficient Nonlinear Loudspeakers—Bo Rohde Pedersen, University of Aalborg - Aalborg, Denmark; Finn T. Agerkvist, Technical University of Denmark - Kgs. Lyngby, Denmark
Loudspeakers have traditionally been designed to be as linear as possible. However, as techniques for compensating nonlinearities are emerging, it becomes possible to use other design criteria. This paper presents and examines a new idea for improving the efficiency of loudspeakers at high levels by changing the voice coil layout. This deliberate nonlinear design has the benefit that a smaller amplifier can be used, which, in turn, has the benefit of reducing system cost as well as reducing power consumption.
[Poster Presentation Associated with Paper Presentation P19-4]
Convention Paper 6778 (Purchase now)
P26-4 A Dipole Multimedia Loudspeaker—Vladimir Filevski, Broadcasting Council of Macedonia - Skopje, Macedonia
A multimedia/computer loudspeaker usually stands on a desk, so the reflected sound from the desk interferes with the direct sound from the loudspeaker. This results in a comb-like frequency response, with first minimum deep at least -8 dB, followed higher in the frequency by a peak of about +4 dB, and so on. This paper describes the design of a dipole multimedia/computer loudspeaker, with less than +2 dB/ -2.4 dB of difference between resultant frequency response (including reflected sound from the desk) and anechoic response.
[Poster Presentation Associated with Paper Presentation P19-7]
Convention Paper 6781 (Purchase now)
P26-5 A Compact 120 Independent Element Spherical Loudspeaker Array with Programmable Radiation Patterns—Rimas Avizienis, Adrian Freed, Peter Kassakian, David Wessel, University of California at Berkely - Berkeley, CA, USA
We describe the geometric and engineering design challenges that were overcome to create a new compact, 10-inch
diameter spherical loudspeaker array with integrated class-D amplifiers and a 120-independent channel digital audio interface using Gigabit Ethernet. A special hybrid geometry is used that combines the maximal symmetry of a triangular-faceted icosahedron with the compact planar packing of six circles on an equilateral triangle ("billiard ball packing"). Six custom 1.25-inch drivers developed by Meyer Sound Labs are mounted on each of 20 aluminum triangular circuit boards. Class D amplifiers for the six loudspeakers are mounted on the other side of each board. Two pentagonal circuit boards in the icosahedron employ Xilinx Spartan 3E FPGA's to demultiplex digital audio signals from incoming Gigabit Ethernet packets and process them before feeding the class-D modulators. Processing includes scaling, delaying, filtering, and limiting.
[Poster Presentation Associated with Paper Presentation P19-9]
Convention Paper 6783 (Purchase now)
P26-6 Constant Directivity End-Fire Arrays for Public Address Systems—Filip Verbinnen, University of Southampton - Southampton, UK
The directivity of current public address systems is controlled very well in mid and high audio frequencies using arrays or horns. Low frequencies, though, are mostly still omnidirectional. The cardioid subwoofer is making its introduction but has some drawbacks limiting the maximum sound pressure level achievable by this type of system. As a possibly better alternative, the end-fire line array is considered as a directive bass system. Some research has already been done on end-fire arrays but none exploited the current potential of digital signal processing techniques. Using a linear end-fire array of loudspeakers each with its own digitally-processed input, the possibilities and limitations of these tapered end-fire linear arrays were examined with the main goal to create a constant directivity end-fire array with a usable frequency range from 20Hz to 200 Hz.
[Poster Presentation Associated with Paper Presentation P19-11]
Convention Paper 6785 (Purchase now)
P26-7 DGRC Arrays: A Synthesis of Geometrical and Electronic Loudspeaker Arrays—Xavier Meynial, Active Audio - St. Herblain, France
Loudspeaker arrays offer an efficient way of achieving both uniform SPL coverage and high sound clarity over a large audience area. Two types of arrays have been proposed over the last 15 years: geometrically steered J shape arrays, mainly for high power sound reinforcement; and electronically steered vertical arrays, mainly for speech diffusion in public spaces. This paper introduces the Digital and Geometric Radiation Control (DGRC) principle, which combines the advantages of geometrical arrays and electronic arrays; an array that is vertical so that it can be mounted on a wall; that is controlled with great flexibility using its DSP; and that the power is evenly distributed upon loudspeakers.
[Poster Presentation Associated with Paper Presentation P1921-12]
Convention Paper 6786 (Purchase now)
P26-8 Universal System for Spatial Sound Reinforcement in Theaters and Large Venues—System Design and User Interface—Frank Melchior, Gabriel Gatzsche, Michael Strauss, Katrin Reichelt, Martin Dausel, Joachim Deguara, Fraunhofer IDMT - Ilmenau, Germany
Sound reinforcement for large venues is a challenging task. Up to now most of the systems and concepts are focused on a more or less stereophonic reproduction. Beside these concepts a promising technology exists, which enables a spatial sound reinforcement for a larger audience. Spatial sound reinforcement is an important aspect especially in high quality applications like opera houses and venues for classical music. This paper presents an innovative system- and multi-user interface concept for dynamic automation and interactive control of sound source positions and other properties for variable reproduction systems in live sound reinforcement applications. The system has been designed in close cooperation with experts in sound reinforcement for opera houses. The developed user interfaces are described in addition to a detailed view on the practical realization and audio processing in such a system.
[Poster Presentation Associated with Paper Presentation P19-13]
Convention Paper 6787 (Purchase now)
P26-9 Sound Field Characterization in Audio Reproduction with the Bit-Grouped Digital Transducer Array—Jorge Mendoza-López, Simon C. Busbridge, University of Brighton - Brighton, East Sussex, UK; Peter A. Fryer, B&W Group Ltd. - Steyning, West Sussex, UK
A bit-grouped digital transducer array loudspeaker with different numbers of nominally identical transducers for each bit has been developed. The direct digital-to-acoustic conversion process produces a sound field whose quality is shown to be spatially dependent and highly influenced by real effects including nonuniform transducer frequency responses, transducer mismatching, and baffle size. Spatial sound pressure and total harmonic distortion maps show that reducing the array size leads to improved reconstruction due to reduced phase distortion. For a given sampling rate and signal frequency, total harmonic distortion decreases as the listening distance is increased. A new criterion for the sweet-spot location in digital arrays is proposed based on the difference between the distortion introduced by path-length differences and the inherent quantization distortion.
Convention Paper 6835 (Purchase now)
P26-10 Radiation Impedance of Transducer Field Driven by Binary Signals—Libor Husník, František Kadlec, Czech Technical University in Prague - Prague, Czech Republic
This paper addresses another aspect of transducers with the direct digital to analog conversion, sometimes called digital loudspeakers, which is the radiation impedance. In a transducer array embodying such a system, every value of the acoustic pressure from its dynamic range is radiated by a different number of elementary transducers, i.e., different total surface of membranes, driven by the PCM signal. Since the critical frequency depends mainly on the total surface of membranes, an interesting phenomenon appears, i.e., every sound pressure level is radiated with a different radiation impedance. As a result, different levels may be radiated differently.
Convention Paper 6836 (Purchase now)
P26-11 An Introductory Review for U-fa (USM Driven Woofer) Development—Hirokazu Negishi, University of Essex - Yokosuka, Japan
The concept of “U-fa” was born twelve years ago. Since that time, there has been a lot of development in that field. This paper was originally intended to be presented at the AES/NY Convention in 2001 but was cancelled for various reasons. However, a second revival of the activity brought about several Japanese convention papers, and they debuted in AES/NY last October. Since the audio world and Ultrasonic Motor has little common ground, it seems difficult to appreciate what actually made the difference by introducing USM to Woofer. In order to bridge the gap, the original paper has been totally rewritten. Emphasis is now put on introductory reading for early day’s and the background of U-fa, rather than on theories and equations.
Convention Paper 6837 (Purchase now)
P26-12 Improved Model of Loudspeaker Using Continuous Revolution of Ultrasonic Motor—Yusuke Iwaki, Yuta Ohnuma, Juro Ohga, Shibaura Institute of Technology - Minato-ku, Tokyo, Japan; Hirokazu Negishi, DiMagic Co. Ltd. - Tokyo, Japan; Kazuaki Maeda, TOA Co. - Kobe, Japan
The loudspeaker using continuous revolutions of an ultrasonic motor (USM), proposed by the authors, is suitable to radiate sound at very low frequency. This paper describes an improved model of the USM loudspeaker. Functions of rotor and stator is more inversed than the first model to simplify the electric connection. Mass of the rotating ring is increased to make the inertial force larger. A silicon rubber joint is used to connect USM and the cone radiator to avoid frictional noise.
Convention Paper 6838 (Purchase now)
P26-13 Ring Element Model: Program Results—Elena Prokofieva, Linn Products - Waterfoot, UK
The new modeling method was described in a series of papers presented at AES conventions in 2004 through 2005. In the proposed model the general approach is to represent a cone driver as a set of rings, loaded by a concentric force, applied around the lower element’s edge. The ring element is preferred to the finite element model due to its simple yet precise driver simulation. The standard theoretical model of the radiating piston was initially considered. The problems inherent to this approach were highlighted, and the model was improved by removing standard assumptions one-by-one and replacing them with more complex calculation procedures for improved simulation. The first group of assumptions is programmed and compared against the real measurements and analytical calculation. The advantages of each model are studied and an explanation of how each of the standard theoretical assumptions affects the final result is provided. The possibilities to use different models in preliminary loudspeaker design are also discussed in this paper.
Convention Paper 6839 (Purchase now)
P26-14 Analysis and Optimal Design of Miniature Loudspeakers—Mingsian Bai, Rong-Liang Chen, National Chiao-Tung University - Hsin-Chu, Taiwan
Miniature loudspeakers are key components to many 3C products especially for portable devices such as mobile phones, PDAs, MP3 players, etc. Due to size limitation, miniature loudspeakers suffer from the problem of low output level. To gain higher output, one tends to drive the miniature loudspeaker over the excursion limit and induce nonlinear distortion. Thus, how to best reconcile the conflicting requirements of nonlinear distortion and acoustic output is extremely crucial in the design of such loudspeakers. To address the issue, this paper presents a systematic procedure to pinpoint the optimal designs appropriate for miniature dynamic moving-coil loudspeakers. The optimization procedure is based on an electro-acoustic model established by using the test-box method. Characteristics including voice-coil impedance, frequency response, and harmonic distortion are evaluated. The results show that significant improvement in output performance and excursion limitation has been gained by using the optimal design.
Convention Paper 6840 (Purchase now)
P26-15 Positions Effect of Multi Exciters and the Optimization on Sound Pressure Responses of Distributed Mode Loudspeaker—Suzhen Zhang, Yong Shen, Nanjing University - Nanging, China; Xiaoxiang Shen, Creative Technology (China) Co., Ltd. - Shanghai, China
The exciters of distributed mode loudspeaker (DML) mainly play two roles, activating forces and attached masses, both of which will affect the sound pressure response of the panel. Therefore, to derive a smoother sound pressure response, the positions of the exciters should be considered carefully. In this paper the model of a panel with activating forces and attached masses is developed with partial differential equations (PDEs) in FEMLAB. The optimized positions of the exciters are given combining the use of genetic algorithm (GA) based on two different optimization criteria, sound pressure response and mode distribution. Optimal results in both cases are derived and show that various optimization criteria lead to different sound pressure responses and sound pressure sensitivities.
Convention Paper 6841 (Purchase now)
P26-16 Simulation of Reconstruction of Oversampled Signals in Digital Loudspeakers—Haihua Zhang, Simon C. Busbridge, Chris Garrett, University of Brighton - Brighton, UK; Peter A. Fryer, B&W Group Ltd. - Worthing, West Sussex, UK
The technique of over sampling and noise shaping has the potential to improve the resolution of digital loudspeaker systems, at the expense of increasing the signal bandwidth. Previous work has shown that the acoustic radiator in a digital loudspeaker system can act as a reconstruction filter if the over-sampled signal bandwidth exceeds the transducer bandwidth. If the oversampled signal is within the transducer bandwidth, the use of reconstruction filters has to be considered in the system. This paper presents an investigation of reconstruction with both pre-acoustic and postacoustic filtering. Mathematical modeling suggests that the reconstruction in a direct digital-to-analog loudspeaker should take place before the summation of the digital bit streams to avoid intermodulation distortion. This is counter-intuitive, because the electronic driving signals are no longer digital in the digital loudspeaker system.
Convention Paper 6842 (Purchase now)
P26-17 Digital Measurement for Dynamic Distortion of Loudspeakers—Keiichi Imaoka, Juro Ohga, Shibaura Institute of Technology - Minato-ku, Tokyo, Japan
Most measuring methods using digital signal processing techniques developed recently are only for measurements in a linear range. There is still no suitable measuring method by a digital processing system for measurement of nonlinear distortion of acoustical devices. An accurate and convenient nonlinear distortion measurement system should be developed. The authors already proposed a new digital distortion measuring method for acoustical devices. This method applies a Pink-TSP signal (Time Stretched Pulse, i.e., quickly swept sinusoidal signal), whose frequency band is partially eliminated, to an acoustical system to be measured. The detected component produced in the rejected band is measured as a distortion. This paper analyzes experimental results by using a single resonant system.
Convention Paper 6843 (Purchase now)
P26-18 Effect of Membrane Damages on Loudspeaker Performance—Romuald Boleiko, Wroclaw University of Technology - Wroclaw, Poland
This paper deals with the effect of membrane damage on loudspeaker parameters. Tweeters with small dents in their metal membranes and with a perforation in their fabric membranes were tested. Acoustic parameters of the loudspeakers were investigated by measuring the sound pressure level frequency response and vibrations of loudspeaker’s membrane. A scanning laser Doppler vibrometer was used for the latter measurements. It was found that dents and a perforation in the tweeter’s dome may change its frequency response by more than 6 dB. As a rule, dents in the dome affect the tweeter frequency response at high frequencies while a perforation in the soft fabric dome affects the tweeter response at medium frequencies.
Convention Paper 6844 (Purchase now)
P26-19 Loudspeaker Testing at the Production Line—Wolfgang Klippel, Stefan Irrgang, Ulf Seidel, Klippel GmbH - Dresden, Germany
Quality control in the mass production of transducers and electro-acoustical systems requires an objective technique for reliable selection of defective units. A new technique is presented for detecting defects that produce almost inaudible symptoms during testing but may degrade sound quality in the final application (e.g., loose particles in the gap). Here the regular distortion that is characteristic for good units is modeled and actively compensated in the measured signal of a device under test to reveal symptoms of irregular defects (meta-hearing technology). This paper shows ways to perform high-speed measurements close to the physical limits and how to cope with ambient noise in a production environment. Traditional and more advanced techniques for separating passed and failed units are compared and their integration into process control is discussed. Finally, the paper addresses cost effective implementation in a robust hardware, flexibility to customer’s needs, simple handling, and other practical requirements.
Convention Paper 6845 (Purchase now)
P26-20 New Structure of Loudspeaker—Guy Lemarquand, University du Maine - Le Mans, France
We present a new structure of loudspeaker: the motor is ironless, the suspension is ferrofluidic, and the moving part is piston-like, with a concave dome. The absence of iron guarantees a small and constant inductance of the moving coil, as well as the absence of eddy currents. The motor includes two circular joints, one on each side of the moving coil. These joints are ferrofluidic. They fulfill the guidance and centering function and the air tightness function. This structure is quite rigid. As there is no traditional suspension in this structure, the related nonlinearities and hysteresis disappear.
Convention Paper 6846 (Purchase now)
|(C) 2006, Audio Engineering Society, Inc.