DAGA 2026
Acoustic Conference in Dresden
The 52nd Annual Conference on Acoustics is being organized jointly by the German Acoustics Society (DEGA), the Technical University of Dresden and the Polish Acoustical Society. DAGA will offer interesting plenary lectures from a variety of acoustics fields and opportunities for creative exchanges with experts.
This year the event will take place from March 23th to 26th, 2026 at the International Congress Center in Dresden.
You will have the opportunity to attend four pre-colloquia, several structured sessions on current topics, technical lectures, and poster presentations. In addition, technical committee meetings and organized excursions will take place throughout the event.
Our KLIPPEL team will also be attending the DAGA. Don't miss out on this event and check back regularly for updates!
➥ Be part of DAGA 2026 in Dresden meet our engineers and join the various demos at our booth T21. Discover more on the official DAGA website!
General Information about the DAGA 2026
Date: March 23-26,2026
Venue: International Congress Center in Dresden
Visit us at booth T21 at terrace level.
Visit here for more information on the program.
Experience the Conference Program
KLIPPEL is represented in the DAGA programme with five presentations.On 23 March, you will have the opportunity to attend four fascinating pre-colloquium sessions. We are delighted to announce that Prof. Wolfgang Klippel and our colleague Eliot Deschang will both be presenting on the topic of "Electroacoustics: Novel Loudspeaker and Microphone Technologies", delivering two insightful paper sessions each. On 25 March, you will have the chance to attend two paper sessions on the topics of “Electroacoustics and Audio Signal Processing” and “Physical Acoustics”, as well as Wolfgang Klippel's plenary session.
"Loudspeaker Modeling with Magnetic Modes, Part 1: Theory" by Wolfgang Klippel, Eliot Deschang
Monday, 23 March 2026 from 13:45 to 14:10 in Conference Room 1
Abstract: Most loudspeakers and other audio devices use a coil, a magnet, and iron to generate the driving force and a back-induced voltage. At small amplitudes, the electromechanical transduction can be described by a constant coupling factor, Bl, voice coil inductance, and DC resistance Re. At higher amplitudes, the force factor Bl(x,i,f) and inductance L(f,x,i), depend on voice coil displacement x and input current i, generating harmonic and intermodulation distortion, and limiting the maximum acoustical output. Due to eddy currents, shorting rings, and hysteresis, the nonlinearities exhibit memory and display nonlinear symptoms that depend on the frequency f. The paper presents a new model that exploits a modal decomposition of the magnetic field to separate the frequency-dependent reluctance in the iron from the static nonlinearities. The magnetic modes improve the accuracy of the modelling while using a reduced number of parameters that have physical meaning. This perspective is beneficial for optimizing the magnetic field, addressing saturation issues in the iron, and placing shorting rings to reduce the inductance. The combination of increased modeling strength with low complexity provides a powerful basis for actively canceling signal distortion generated by the motor nonlinearities over the audio bandwidth.
"Loudspeaker Modeling with Magnetic Modes, Part 2: Identification and Interpretation" by Eliot Deschang
Monday, 23 March 2026 from 14:10 to 14:35 in Conference Room 1
Abstract: Magnetic mode decomposition provides an efficient framework for analyzing the electromagnetic behavior of loudspeaker motor structures across different topologies. The input data for identifying these modes can be obtained either from finite element simulations or from point-by-point measurements of the electrical input impedance at various frequencies with the voice coil clamped at fixed positions. For moving-coil transducers, the modal parameters are extracted using Non-negative Matrix Factorization (NMF), which separates the frequency-dependent and displacement- dependent contributions of the input dataset. In practice, two to three modes are sufficient to reproduce more than 80 % of the original information, offering a compact yet physically interpretable representation. This approach enables more accurate modeling of the loudspeaker inductive behavior and facilitates diagnostic and optimization tasks for loudspeaker motor design.
"Validating Finite Element Simulations of Nonlinear Inductance and Force Factor of Electrodynamic Transducers with Measurements" by Jonathan Gerbet
Wednesday, 25 March from 9:00 to 9:20 in Hall 5
Abstract: Finite Element Analysis (FEA) is a powerful tool for predicting the behavior of electro-dynamic transducers without the need for numerous physical prototypes, significantly reducing both cost and development time. However, validating FEA results remains challenging and time-consuming. In this paper, a fast, full dynamic, and non-destructive measurement technique is applied to validate FEA results of the key parameters of the nonlinear transducer motor: the position-dependent force factor (Bl(x)) and the frequency- and the position-dependent lossy self-inductance. The study compares the nonlinear motor parameters obtained from both measurement and simulation. It further identifies which of these parameters are most critical for accurately predicting transducer behavior.
"Grey box modeling - Learning with physics" by Wolfgang Klippel
Wednesday, 25 MArch 2026 from 11:45 to 12:30 in Main Hall
Abstract: For 100 years, almost all loudspeakers have used the same electrodynamic transducer principle. Loudspeakers are inefficient, cause audible signal distortion and are therefore the weakest link in the transmission chain. However, telecommunications, the automotive industry and other mobile and stationary applications, such as home entertainment systems and professional sound reinforcement systems, increasingly require small, lightweight and energy-efficient transducers that can be mass-produced at low cost. In all these applications, the loudspeaker is neither linear nor time-invariant. Electrical, mechanical and acoustic nonlinearities, heating, material fatigue and external influences (e.g. weather) limit sound pressure and impair sound quality. This lecture reports on progress in loudspeaker modelling over the last 40 years, particularly with regard to large-signal behaviour. New grey-box models have been developed that use existing a priori knowledge, such as fundamental physical relationships, to form structures and optimally estimate free model parameters using measured data. These models are now indispensable for the practical development and diagnostic evaluation of loudspeaker systems, forming the theoretical basis for adaptive, nonlinear control that can be cost-effectively integrated into digital amplifiers.
"Cosine Expansion and Fourier Transform of Radial Displacement Fields on Circular Loudspeaker Membranes" by Ronald Starke
Wednesday, 25 March from 17:20 to 17:40 in Conferenz Room 2
Abstract: We discuss the Fourier series expansion of a radial function on a finite disc obeying a Dirichlet boundary condition at the outer edge. This is a typical situation for the displacement field of a loudspeaker membrane. Concretely, we first argue that such a function has a natural expansion in terms of cosine functions defined with nonstandard wavenumbers. Subsequently, its relation to the Fourier transform of a symmetrically extended radial function is discussed and illustrated with practical examples.
