ABSTRACT. A small number of ethnic instruments from different parts of the world produce a single fundamental frequency in their traditional use. Most notable are the didgeridoo, the mouth-harp, and the overtone singing technique. Constraining a musical instrument to play a single frequency sounds like a contradiction in terms, however these instruments resort to a variety of means to nevertheless make their sound rich and musically interesting. In this talk we will discuss the characteristic spectra of these instruments, how this influences their playing styles, and how they rely on the perception of the listeners to create music that is pleasing to traditional as well as contemporary audiences.

ABSTRACT. Wine glasses can be used to produce musical sounds, either as standalone makeshift instruments or as components of carefully tuned instruments such as the Glass Harp. Typically, wine glasses are played either by rubbing a moistened finger around the rim, or by using a mallet or a bow. The glasses are tuned to specific pitches and each glass is used for producing a single note. The current study proposes a new method for exciting wine glasses, using a concept similar to that of sympathetic strings, found in such instruments as the Sitar. While the Sitar uses sympathetic resonance between sets of coupled strings, this research explores the idea of using sympathetic resonance between strings and wine glasses, where the vibration of the string produces vibration in the glass. Two methods for creating string-glass coupling were explored: 1) by direct contact; 2) by using an intermediate component similar to a string instrument's bridge. The coupling component was developed by a process of topology optimization using FEM simulations, resulting in a small cane-shaped design. Experiments tested the responses of three wine glasses having different frequency responses to excitation by pitched strings using both methods.The findings show that both methods produce audible responses from the glasses, with the direct string-glass contact method having the loudest responses. Each glass responded to several different pitches, with each response having a different spectrum. The methods developed in this research offer a new approach to the design of wine glass based instruments. Instruments designed with this approach can have a small number of glasses, each producing different notes without being directly excited by the player. The use of the coupling component allows for greater flexibility in the placement of both glasses and strings.

ABSTRACT. The Himalayan singing bowl is a nearly symmetric metal bowl that produces a musical sound by rotating a wooden stick around the outside rim. The wooden stick, referred to as a puja, excites vibrations through a stick-slip mechanism. The amplitude of vibration is directly related to the applied force and velocity of the puja as it moves around the bowl. Typically only a single vibrational mode is excited, resulting in narrow-band oscillation at a frequency determined by the size of the bowl. However, as the angular velocity of the puja increases and/or the force of the puja on the bowl decreases, an audible chatter will often occur. The goal of the work reported here is to determine the origin of this chatter. Initial results indicate that the puja does not lie directly on a node of the radiating mode. Furthermore, the position of the node with respect to the puja changes as a function of rotational velocity. Because the puja does not lie directly on a node, it appears that as the amplitude of the radiating mode is increased the puja can lose contact with the bowl resulting in a chattering sound.

ABSTRACT. For musical wind instruments, the problem of the transition between a cylindrical tube and a conical one occurs in many situations. In a sufficiently long cylinder, only planar waves can propagate, while in a sufficiently long cone, only spherical waves can propagate. Modelling the matching between the two tubes requires the consideration of higher-order modes, which are evanescent, i.e., close to the matching volume. A mode-matching method yields complicated computation, thus a numerical discretization method (such as the Finite Element Method, FEM), or experiment are more suitable. However the general shape of the model is known, under the condition that only one mode propagates in each tube, i.e., that the frequency is low enough. The model includes the effect of compressibility of the matching volume, and three acoustic masses. The major result is that the latter are linked together by the geometrical parameters, therefore only one mass has to be determined numerically or experimentally. It is proportional to the so-called length correction. Its knowledge allows the determination of an analytical model valid up to the first cutoff frequency. The results of the analytical formulas are compared with those of the FEM method, and the agreement is very good. The literature on this subject is rare, except the papers by Chester (J. Sound Vib. 1987) and Martin (J. Eng. Math., 2011). Comparison with formulas found in these papers is carried out, and it is found that the present study improves their results. Furthermore, for the particular case of a cylindrical tube radiating with an infinite flange, the analytical formula compares very well with the exact result (Norris and Sheng, J. Sound Vib. 1989), up to the first cutoff frequency. This work was financed by the research project CAGIMA supported by the French Agence Nationale de la Recherche.

ABSTRACT. The aim of this work is to highlight experimentally how inharmonicity of the bore res- onance frequencies of an alto saxophone influence the nature of the oscillation regimes.

A variable volume branching from the neck of an alto sax at an appropriate position allows to change the frequency of the first resonance independently from the second. A blowing machine with artificial lips is used to make the saxophone play while controlling independently the control parameters: the blowing pressure and an embouchure parameter. Values of these parameters are estimated experimentally through the measurement of the nonlinear characteristics linking the mean air flow blown into the instrument to the static pressure difference across the reed. Experiments with different values of the control parameters as well as of the inharmonicity produce different kinds of oscillation regimes. These regimes are categorized through the analysis of the pressure signal inside the mouthpiece. The resulting maps demonstrate that the emergence of quasi-periodic regimes, and their extent, depend on the level of inharmonicity, but also on the values of the control parameters. Periodic regimes playable by choosing appropriate values of the control parameters also differ according to the level of inharmonicity, a higher inharmonicity facilitating the emergence of the third register.

Trends highlighted experimentally in this article are in agreement with numerical results obtained in a previous work. This numerical study was based on the use of a minimal model of wind instruments, so it is possible to deduce that features neglected in this model are not determining in the production of quasi-periodic regimes (inertia and damping of the reed, resonance modes higher than the third one)

ABSTRACT. This paper investigates the correlation between input impedance measurements and the played intonation on four Eb-clarinets (german system) in the altissimo register. The experiment con- sisted of an input impedance measurement and a performance recording measurement. 12 main fingerings of the altissimo register (C#6 – C7) were measured and recorded. Furthermore, 56 alternative fingerings were investigated. The input impedance measurement showed that the intonation of the main fingerings in the altissimo register increased with higher notes and the impedances peaks decreased for all four clarinets. We also ob- served two notes on all clarinets (G#6, B6) where the main fin- gerings resulted in too low intonation (-20–30 cent). The perfor- mance measurement confirmed this observation and allows to subdivide the altissimo register into 3 ranges. In the first range the main fingerings resulted in correct intonation. In the second range alternative fingerings helped to correct the pitch and in the third range, correct tones were impossible to play. One rea- son for this might be the reed’s natural frequency, which is in the frequency range of the altissimo register on the Eb-clarinet. Other possible influences to the intonation may be the player’s vocal tract setting and the setting of the embouchure.

ABSTRACT. Electromechanical and electroacoustical analogies are widely used in mechanics and acoustics. Highly developed theories for electric networks, including sources, inductive, capacitive and resistive elements, as well as resonating circuits, electrical lines, etc. can thus be directly used to simulate the behaviour of musical instruments.

There exist special Circuit Analysis Programs, such as Micro Cap [1], that are useful to simulate complex electrical networks. Applying the above mentioned analogies, these tools can be used in different ways to simulate characteristics of wind instruments, such as input impedance, transient behaviour, spectra, etc.

The authors have developed special so called “macros” that allow a convenient application of the software tool for mechanical and acoustical systems.

A method was developed to simulate a complete clarinet body (without reed and mouthpiece). This method was then used successfully to develop a new G-high clarinet, a special Viennese instrument, called Picksuesses Hoelzl and used in Schrammelmusik. How this was done, also with the help of BIAS [2] will be described in detail in the article.

The purpose of this paper is to show the possibilities that exist, using software-based models of wind instruments, especially clarinets. It is not meant to be a recipe for instrument manufacturers. Intensive training would be needed to use the described method in practice.

ABSTRACT. Obtaining data on the directional characteristics of musical instruments can be both time-consuming and demanding in terms of equipment. This paper explains the construction of an affordable automatic turntable system, for measuring the radiated sound as a function of angle and frequency in one plane. An initial test was carried out on two brass instruments excited by an artificial mouth and two stringed instruments driven by a shaker. The second part of this paper deals with the further processing of the obtained values. As the sound pattern data can be stored in an online database (MySQL), the directional characteristics are accessible via a web-interface. The user is able to display the measurements as single plot (e.g. for each angle), as polar-plot or colormap. This interface represents the first part of a planned Musical Instrument Measurement Database (MIMD) for collecting diverse measurement data of different kind of musical instruments.

ABSTRACT. This contribution presents bassoon blowing experiments with an artificial mouth. Given the precise adjustability of lip position and blowing pressure, the challenge in the presented experiment is to find those parameter combinations, that produce "reasonable" sustained sounds. In a large set of blowing experiments, artifical mouth adjustment parameters are identified, for which sounding pitch is within 5 Cents deviation from the nominal pitch expected for that fingering. For those musically relevant regimes, an overview on the steady-state parameters of bassoon playing is given. The overview comprises artificial mouth parameters (lip force, blowing pressure, reed slit height, time-averaged flow-rate, RMS reed pressure, and reed pressure shape parameters) and sound related parameters (harmonic spectral centroid, formant frequencies, and loudness). The experiments were carried out with one and the same synthetic bassoon reed on five modern german bassoons and cover the complete playing range in frequency (notes) and dynamics (loudness). The observed variability between the five instruments is relatively small. Yet the comparison of the steady-state parameters with acoustic impedance curves provides some insight into the resonator-reed interaction in the bassoon.

ABSTRACT. Numerical sound synthesis is often carried out using the finite difference time domain method. In order to analyse the stability of the derived models, energy methods can be used for both linear and nonlinear settings. For Hamiltonian systems the existence of a conserved numerical energy-like quantity can be used to guarantee the stability of the simulations. In this paper it is shown how to derive similar discrete conservation laws in cases where energy is dissipated due to friction or in the presence of an energy source due to an external force. A damped harmonic oscillator (for which an analytic solution is available) is used to present the proposed methodology. After showing how to arrive at a conserved quantity, the simulation of a nonlinear single reed shows an example of an application in the context of musical acoustics.

ABSTRACT. Proponents of the Pulse Forming Theory (for instance Fricke and Voigt) claim that the reed closing time of wind instruments remains approximately constant over their playing range. The theory presented in \cite{Ollivier02} might provide an explanation for this phenomenon in terms of the geometry of the resonator. This paper aims to test the hypothesis that, for a conical instrument, the closing time of the reed is proportional to the truncation length of the cone. This is done by simulating (through physical modelling) and recording an oboe with a normal staple and with a short staple, assuming that the staple at the top of the instrument is part of the resonator. While the simulations confirm the hypothesis, the recordings show interesting results that invite us to postulate other possibilities.

ABSTRACT. This paper investigates the satisfaction of the intonation on the Eb-clarinet. The participants (N = 39, male = 27, women = 12) had to judge the intonation in different registers on their Eb-clarinets. The tones (E3 – C7) were subdivided into the chalumeau register, the throat tone register, the clarion register and the altissimo register. Furthermore, I calculated the mean values for the satisfaction of the intonation of all participants. Overall, four groups joined this survey: professional players, advanced students, beginning students and sideline musicians. The questionnaire showed that all groups reported having problems in the throat tone register and the altissimo register.Furthermore, the satisfaction with the intonation of the professional group (48%), which is able to judge the quality of instruments more accurate than other groups, was observed. The professional group reported having problems only in the altissimo register.

 

ABSTRACT. Arthur Benade introduced the notion of a tone-hole lattice in the early 1960s [1], [2]. He found, that there exists a so-called “cutoff frequency” that is determined by the structural dimensions of the tube and its side-holes. Since then several other researchers have studied the properties of the tone-hole lattice, especially dealing with the row of open tone-holes. When simulating mechanical and acoustical systems it is very convenient to use electromechanical and electro-acoustical analogies. Highly developed theories for electric networks can thus be directly used to simulate the behaviour of musical instruments. Special software applications, such as Micro-CAP [1], a Circuit Analysis Program, can be used for this purpose. To build a bridge from the electrical to the mechanical world there exist so called "macros" that were developed mainly by the second author, with mechanical input parameters, so the user need not necessarily think in electrical terms. Examples of such macros are two-ports representing lossy cylindrical and conical tubes, and two-poles for short holes etc. which occur in wind instruments. Other two-ports are ideal transformers for coupling mechanical and acoustical worlds via the connecting area. The applicability of the simulation tool for the tone-hole lattice was tested using the work of A. Benade as an entry point. In his famous book Fundamentals of Musical Acoustics, Benade gave an example of a tube 61 cm long extended by a tube with side holes. The result is shown in Fig. 21.3 of the book. No more details about the tube are given there. Using the simulation tool, the authors were able to reconstruct approximately the dimensions of the tube Benade used for his measurement. The influence of the cutoff frequency can be seen clearly. Not much is said in the literature about the regularities of the resonance peaks above the cutoff frequency. The impedance-versus-frequency diagrams that are easily derived with the aid of Micro-CAP can help much to detect influences of the several dimensions of the tone hole system. It is also possible to show pressure and flow profiles along the axis of the tube with opened and closed side-holes. Such work can lead to a further understanding of the properties of real woodwind instruments.

ABSTRACT. An idiosyncratic feature of all lute instruments is the specific fixation of the strings leading to certain physical boundary conditions. Alongside the structural functionality of the string adjustment, the bridge and the nut, fixing the taut string at both ends, the string fixation transmits vibrational energy from the strings to the instrument's resonator and vice versa. An effect arising from this bi-directional coupling is a moving string fixation leading to time varying boundary conditions. In addition to this, the specific fixation of the strings leads to differing admittances in the respective transverse polarization of the string, giving rise to non-linear energy transfer between the polarizations. In this treatise, several fixtures are modeled, showing the influence of the boundary conditions on the string motion as well as the mechanical coupling strength in the respective direction of motion. The strings are modeled using finite difference approximations for the spatial domain and a locally symplectic integration scheme for the time domain discretization. The numerical results are compared to measurements taken with a high-speed camera and a piezoelectric transducer, giving insight to the feasibility of the proposed model parameters and showing the influence on the resulting sound.

ABSTRACT. The Brazilian guitars are originally countryside traditional instruments played in different regions of Brazil where several variations of body shapes, types of wood, numbers of strings and tunings are normally found. The present work is focused on the viola capira (portuguese for countryside guitar), which is the most common type and plays a substantial role in traditional and recent Brazilian music. In general, it has ten steel strings arranged in five pairs which are coupled to the soundboard through the bridge in the same way of classical guitars. In order to analyse the interaction between the body and the strings of the instrument, a physical modelling based on an hybrid approach is developped; body modes are identified using experimental modal analysis and are coupled to the modes of an array of strings. A set of time-domain simulations is performed in order to reveal some specificities of the string-body coupling on the Brazilian guitars in terms of sympathetic resonances and beating phenomena.

ABSTRACT. Tanpura string vibrations have been investigated previously using numerical models based on energy conserving schemes derived from a Hamiltonian description in one-dimensional form. Such time-domain models have the property that – for the lossless case - the numerical Hamiltonian (representing total energy of the system) can be proven to be constant from one time step to the next, irrespective of any of the system parameters; in practice the Hamiltonian can be shown to be conserved within machine precision. Models of this kind can reproduce a jvari effect, which results from the bridge-string interaction. However the one-dimensional formulation has recently been shown to fail to replicate the jvari’s strong dependence on the thread placement. As a first step towards simulations which accurately emulate this sensitivity to the thread placement, a two-dimensional model is proposed, incorporating coupling of a controllable level between the two string polarisations at the string termination opposite from the barrier. In addition, a friction force acting when the string slides across the bridge in the horizontal direction is introduced, thus effecting a further damping mechanism. In this preliminary study, the string is terminated at the position of the thread. As in the one-dimensional model, an implicit scheme has to be used to solve the system, employing Newton’s method to calculate the updated positions and momentums of each string segment. The two-dimensional model is proven to be energy conserving when the loss parameters are set to zero, irrespective of the coupling constant. Both frequency-dependent and independent losses are then added to the string, so that the model can be compared to analogous instruments. The influence of coupling and the bridge friction are investigated.

ABSTRACT. This paper addresses the time-domain simulation of beams made of materials that mutate according to the dynamics. The physical model is based on a standard linear Euler-Bernoulli equation, combined with parameterized nonlinear damping models inside a class that preserves the eigen-modes of the conservative problem. Typical mutations between metal, glass or wood are achieved through a local-in-time interpolation of damping characteristics with respect to the energy. This results in a structured nonlinear beam model which proves to be passive. It is recast in the formalism of Port-Hamiltonian Systems, which naturally provides a power-balanced decomposition into conservative, dissipative and source parts and from which a guaranteed passive simulation can be derived. The paper is organized as follows. First, the linear Euler-Bernoulli model of a beam is recalled. A class of linear damping models that preserve the eigen-modes is proposed. The power balance of this model is derived. Second, a modal decomposition is derived, a pole analysis is presented with a special focus on the dependence on the damping parameters. Third, this linear model is recast in the formalism of Port-Hamiltonian Systems: both the infinite-dimensional problem and the finite-dimensional approximation based on a mode truncation are examined. A guaranteed passive simulation and numerical results are provided. Fourth, the extended class of nonlinear damping models based on interpolating functions is introduced. The consistency with the (linear) modal decomposition is examined. The nonlinear Port-Hamiltonian System is provided. Finally, numerical results for various configurations of mutating characteristics are examined, together with sounds.

ABSTRACT. Time-domain simulation of musical instruments has shown promising results in recent years. Particularly attractive from a sound synthesis perspective is the resolution of system displaying some degree of nonlinearity, because of the richness of the perceptual information that nonlinearities produce. In this work, the focus is on one such system, namely a flat (circular or rectangular) plate which, to a first approximation, can efficiently simulate the sound of a gong. From a dynamical perspective, in spite of very different geometries, plates and gongs behave similarly, meaning that the degree of nonlinearity is set by how large the amplitude of vibrations of the flexural waves is. In particular, plates and gongs may attain linear, weakly nonlinear and strongly nonlinear regimes when the amplitude of vibrations is, respectively, much smaller, of the same order of and larger than some defining thickness parameter. The dynamics of plates is well described by a set of two coupled Partial Differential Equations (PDEs) known as the von Kármán equations. For rectangular plates, a family of conservative Finite Difference schemes was developed by Bilbao. An alternative approach is offered in this work, where the von Kármán equations are discretised along the modes of the system in order to reduce the original PDEs to a set of coupled Ordinary Differential Equations (ODEs). This is approach is referred to as modal approach, and it used for here in the context of sound synthesis of nonlinear systems. Salient features of this approach include - implementation of complex decay ratios with no extra effort using modal damping; - simulation of circular plates without bothering with the problems related to particular spatial grids (a frustrating aspect for Finite Difference schemes); - fast computational times for linear and weakly nonlinear regimes.

This work intends to show that the modal approach could be applied to a large class of nonlinear problems, against the common misconception that modes are only useful in treating linear problems. Sound examples and videos are shown in order to complete the presentation.

ABSTRACT. In the last years many new technologies are used in daily musical instrument teaching and learning. Starting with the use of computers and coming to mobile phones and other technologies. In this paper, we try to show possibilities how further new technologies could be used in performance research and the musicians’ daily routine. Starting from data acquisition with common audio and video technologies and several sensor technologies, describing apps for data recording, storage and management up to server based data distribution and visualization and ending with several possibilities of feedback generation. The system is a further development of PART (performing arts technologies) including, but not limited to, the following instruments: Stringed instruments, clarinet, trumpet, trombone, drums and piano. For each musical instrument and problem statement a specific modular set of sensors is integrated to obtain relevant data. These sensor setups are combined with a data logging app and server based online services. A web browser based user interface allows easy access to the data and annotations, data mining techniques for analysis or simple play back e.g. of several vide, audio and data streams together, can be applied. The last point is the problem of providing the right type of feedback in real-time or in distance teaching scenarios. This latter will be shown in some exemplary applications or demos.

ABSTRACT. For a musician’s career it is important to protect own hearing. Professional products therefore should always be custom made. There are several brands doing so called „linear filtering“ by providing band pass filtering for a standard ear channel resonance at 2,7 kHz. They shall now damp the whole frequency range in a regular way, so that sound quality can be preserved while hearing is protected. But: looking at the individual acoustics of an ear channel resonance with its deviation of almost one octave at the peak, the damping can theoretically never be really regular. That includes sound discoloration, which is – in most cases - not beneficial for musical purpose. Several validating tests for hearing protector devices will be presented and referenced to the belonging standards. Every validation test has its pros and cons. To eliminate some disadvantages, a study with a miniature microphone in the ear channel shows results on eight individual ears. The results are some aspects of the impact of the damping with regard to music, which can now be heard because of the recording by the microphone in the ear channel. These results are challenging for audiologists adapting custom made hearing protection for musicians. Individual solutions with different filtering of the two ears are sometimes demanded from certain instrumentalists like flutists or violinists. Regular damping is not always what a musician is looking for: a clarinetist next to a piccolo player for example will be happy getting more damping in the higher frequency range. And last but not least: regular verification as proposed from the industry (leak test and damping verification at 500 Hz) fail completely in music application, where subjective test methods are most advantageous, but need profound expertise.

ABSTRACT. Given the almost universal assumptions in brass pedagogy that good posture allows freer air flow and reduces unnecessary tension, corresponding improvements in sound, flexibility, and articulation may be expected with improved posture. Published studies of posture in brass players, however, have not included systematic observations of the perceived quality of these aspects of performance. Recent work by Friberg and Hunsaker has explored this area of interest, finding significant results with short-term postural training. The neuromechanic of posture is the foundation for understanding efficient and effective movement. The biomechanics, and more importantly, neuromechanics and their role in effective motor control, will be explained and demonstrated in a practical and usable manner for performers. This presentation begins with background information and a brief review of relevant research to be followed by interactive sessions with audience participation.

ABSTRACT. A dual electronic speckle-pattern interferometer (ESPI) system is described that allows images of operational deflection shapes of two sides of a vibrating object to be viewed simultaneously and recorded. Experimental details of the system are discussed and applied to the study of musical drums with two heads at either end of a cylindrical shell. In particular, methods for determining the degree of coupling and the phase relations between the two oscillating heads are shown. A variable length drum has been constructed and used to investigate the coupling of membrane vibrational patterns as a function of the distance between the heads. It is shown that the coupling vs. length data depend strongly on the shapes of the vibrational patterns. These coupling trends are illustrated and further interpreted with the use of a finite element model of the drum, which shows the role of the enclosed air motion within the shell.

ABSTRACT. Electronic speckle pattern interferometry (ESPI) has been implemented for investigation of the vibrational behavior of the Brazilian cuíca. An advanced filtering and processing of time-averaging ESPI data has been applied to a friction drum, which has not been studied by means of laser interferometry before. Asymmetry in the operating deflection shapes of the cuíca has been observed and discussed. The obtained results were compared to finite element method computer simulations and single-point laser Doppler vibrometer deformation measurements. Vibrational shapes of the cuíca occurring under harmonic excitation are presented along with corresponding simulated mode shapes followed by the discussion of irregular patterns. The results of ESPI show an agreement with simulations and provide quantitative data on absolute values of the deformation amplitude.

ABSTRACT. Measuring the radiating sound pressure at surfaces and sound holes of musical instruments can be performed using microphone arrays recording the radiated sound field around the instrument and back-propagating the measured pressures to the radiating surface or holes. Among the advantages of such a measurement setup are the preservation of the radiating surface by not attaching piezos or the like adding additional mass, the possibility to measure both, surfaces and sound holes at the same time, or when using a reasonable amount of microphones with high sampling rate to display transients at sound onsets. The main problem although is the back-propagation itself as this is an ill-posed problem. The main methods of multipole methods like the Minimum Energy Method (MEM), Helmholtz-Lease-Square (HELS) methods or Acoustic Holography (AH) try to overcome this ill-posedness using different approaches which again have pros and cons. While AH uses an angular spectrum back-propagating this spectrum with respect to its phases it has problems with evanescent waves which might appear with low frequencies already. HELS is using one virtual radiation point with many spherical harmonics overcoming the evanescent problem but with poor convergence and the need of very many harmonical functions. MEM uses as many virtual monopoles sources on the surface as there are microphones also overcoming evanescent waves with the need to distort these monopoles to overcome the ill-posedness of the problem. The paper shows examples of these methods for stringed, bowed and wind instruments.

ABSTRACT. Inharmonic components due to body modes are present in the transient phase of classical guitar sounds. The aim of this paper is to investigate their emergence using high resolution analysis methods. Using a wire breaking technique, the near field radiated pressure is recorded in various plucking conditions on classical guitars. An analysis-synthesis approach making use of the ESPRIT method is presented and some indicators of body-sound emergence are proposed and calculated. The influence of the conditions of excitation on the body-sound emergence is first characterized. In a second step, the calculation of these indicators of emergence is performed over a large pool of instruments, including high-end hand-made guitars, as well as entry-range industrial instruments. The main result show an overall greater sensitivity of high end guitars to the excitation conditions with regards to the emergence of the body sound at tone onset, hence allowing an objective categorization of the instruments.

ABSTRACT. A cello bowing pendulum for precise measurements of physical parameters of bowing is presented. It is designed to hold and play bows of any style and to perform the bowing on a strict line. Such a strict line follows the bowing paradigm of musical play but also eases the instrumentation of the applied bowing forces. Two eccentric suspensions translate the initially circular track of a gravity pendulum into the desired straight line. This height-compensated pendulum is additionally weight-compensated by means of a counter-pendulum so that the total system is finally free of any restoring forces. An adjustable fraction of the weight serves as bow force, while the stringed cello is resting beneath the pendulum on a weighing scale which measures this bow force. The tractive force for bowing works strictly perpendicular to the bow force. Both forces in their orthogonal arrangement can be adjusted and measured with centinewton precision. Secondly, bow speed is not predefined by a driving entity but is controlled by the slip-stick interaction. A potential tractive force is provided in combination with a damping unit such that the resulting mechanical impedance will instantaneously adapt to what the actual slip-stick process recommends including a resulting bow speed. This adaptive operation is similar to what a musician senses during bowing and therefore appears to be a valid approach for related performance-based studies of musical acoustics. Such adaptive operation also appears to be preferrable for investigations of bifurkation regimes, of reciprocal inter-instrument pitch-synchronization, physical parameters of slip-stick interaction and of maximum and minimum bow force. Measured signals are organized in multiple data-tuples (every 12 ms of bowing) in order to facilitate statistical analyses and traces. The paper contains the pendulum construction principle, operational modes and ranges, as well as first results from maximum bow force measurements.

ABSTRACT. This treatise is concerned with a measurement method and current results of an on-going project performed at the Institute for Systematic Musicology in Hamburg, Germany. A series of measurements are taken on four concert grand pianos in eight different stages of production, starting with the glue-laminated soundboard planks and ending with the completely assembled piano in concert tuned state. The majority of the sound energy radiated by a grand piano originates from the soundboard, which amplifies the vibrations of the strings via the bridge. Due to the large size of the soundboard as well as its irregular shape, measuring deflection shapes is a nontrivial task. Common measurement methods such as piezoelectric accelerometers can affect the acoustic vibrations of the soundboard due to the added mass. To this end, a noninvasive microphone array method is utilized for the present work. The array consists of 105 microphones successively placed in 18 positions parallel to the soundboard, resulting in a total number of 1890 measurements of which 1289 microphones cover the actual surface. The Soundboard is excited using an acoustic vibrator at 14 positions associated with string termination points on the bass and main bridge. The utilised SineSweep technique is capable of separating the linear part of an impulse response of a weakly non-linear system from several non-linear parts, representing the harmonic distortion of various orders. The measured sound pressure is back-propagated to the radiating soundboard surface using a minimum energy method. The resulting deflection shapes due to the forced oscillation can be visualized up to 4 kHz. Impedance curves are calculated for the 14 input locations on the bridge. The empirical findings will contribute to the formulation of a real-time physical model to help piano makers estimate the impact of design changes on the generated sound.

ABSTRACT. This paper presents an overview of recent research on reed chamber resonances in free reed instruments. Western free reed instruments such as the accordion, harmonica, and harmonium do not normally employ pipe resonators to determine the pitch, but all do feature some sort of reed chamber or cavity in which the reed is mounted. This is necessary to provide a secure mounting for the reed and to properly direct the airstream. The reed chamber will necessarily have resonances which can affect the tone quality and may have some effect on the pitch. Since the cavity volumes are small, however, the resonances will have high frequencies, and the effects on the reed vibration generally tend to be small. An exception to this can occur in the accordion or harmonica for higher pitched reeds, for which a resonance of the reed chamber can be close to the vibration frequency of the reed tongue. In this case the cavity air vibration can become large enough to influence the self-excitation mechanism, possibly interfering with tongue vibration and the resulting musical tone, and in some case preventing the sounding of the reed at all. Builders typically attempt to alleviate this situation by modification of the reed chamber. In the harmonica, if the effect is not too great, skilled players, already accustomed to pitch bending, may be able to overcome this difficulty in some cases by appropriate changes in the vocal tract. In the case of the accordion, Tonon has recently described and implemented means of modifying the internal construction to include a player-controlled internal resonating chamber of variable frequency to enable pitch-bending by the player somewhat similar to that available to the harmonica player.

ABSTRACT. 3D printed vocal tracts modelled from MRI scans of an expert musician performing note bends on a diatonic harmonica are shown to precisely reproduce the same pitch changes. The geometry of the vocal tract acoustically coupled to the reeds is now investigated as an exterior replica or even used as a prosthetic to demonstrate a difficult musical technique. This process is discussed within the history of vocal tract models, and interrogated through a critique of Norbert Wiener’s contentious claim that ‘the best material model for a cat, is another cat, or preferrably the same cat’. We finish with a discussion of how this tool chain of magnetic resonance imaging and numerically controlled fabrication affords novel research in performance techniques through physical simulation.

ABSTRACT. Western free-reed musical instruments include various families: accordions, concertinas, reed organs and mouth organs. In the accordion, each free reed is riveted on a metal plate, named “reed plate”. The reed plate is normally made of aluminium and the reed tongue is a piece of steel. The non-riveted end of the reed is free to vibrate from one side to the other of a slot carved on the reed plate under the reed. These boundary conditions are similar to those of a cantilever beam.

An accordion reed is only activated when air driven by the bellows comes from the same side of the plate on which the reed is riveted (inward-striking reed or inward-sliding reed). As it has little damping, the reed vibrates with a frequency just below the lowest resonance frequency of the clamped-free bar. Apart from this, due to the non linearity of the problem, a great number of harmonics are found for that frequency instead of the non-harmonic overtones expected for a clamped-free bar. On the reverse of the reed plate there is another identical reed with its corresponding slot. One of the reeds sounds when opening the bellows and the other reed sounds when closing the bellows.

In the case of medium or large reeds, the opposite side of the slot is completely covered by a strip made of leather or plastic that moderates the airflow and avoids the passage of air through the hole in the reed tongue that it is not activated.

Bigger reeds produce lower notes. A small mass is placed on the tips (free end) of the accordion's lowest reeds to attain low frequencies without being too long. Accordion reeds are usually tuned by removing material at the free end (to raise the pitch) or removing it in the middle (to lower the pitch). Reed tongues must be carefully contoured: a sharp change of section should be avoided as this is liable to cause a “stress raiser” which can result in a fracture forming.

Accordion players distinguish between bellows attacks and finger attacks. In bellows attacks the button (or key) is pressed first and the bellows are moved after. In finger attacks, the bellows are set in motion (by pulling or squeezing them) and soon afterward the button is pressed down.

In this work, we present the relationships between the control of a concert accordion, the generated sounds, and how these are perceived are analysed. We explore the mass load effect on sound timbre and the fine differences in timbre for bellows attack and finger attack.

ABSTRACT. Attack transients of harmonium-type free reeds from American reed organs have been studied in some detail. Oscillation waveforms were obtained using a laser vibrometer system, variable impedance transducer proximity sensors, and high speed video with tracking software. Although the fundamental transverse mode is dominant, the presence of higher transverse modes and torsional modes in the initial transient has been established. Typically the motion of the reed tongue begins with an initial displacement of the equilibrium position, often accompanied by a few cycles of irregular oscillation. This is followed by a short transitional period in which the amplitude of oscillation gradually increases and the frequency stabilizes at the steady state oscillation frequency. In the next stage, the amplitude of oscillation continues to increase to the steady state value. Spectra derived from the waveforms in each stage have been analyzed, showing that the second transverse mode and the first torsional mode are both observed in the early stages of the transient, with the torsional mode often appearing earlier. Measurements on reed tongues of different design have been made to further explore the significance of the torsional mode in the initial excitation. [Work partially supported by United States National Science Foundation REU Grant PHY-1004860.]

ABSTRACT. In these interactive sessions, posture and effective motor control will be explained and demonstrated in an accessible, hands-on manner for performers and teachers. Recommendations are made for identifying, examining, and planning an intervention to improve dynamic posture. Brass players are invited to bring their modern or period instruments. Other instrumentalists and vocalists are also welcome to participate.

Participants and observers in this two-part session will

1. Observe the interaction between the biomechanical and neuromechanical components of dynamic performance;

2. Recognize components of an efficient static posture;

3. Utilize principles of applied physiology, biomechanics, and motor control to modify posture;

4. Apply a rubric for identifying biomechanical and neuromechanical dysfunctions of posture and designing an individual program for performers.

5. Receive an individualized plan to improve dynamic posture in their performance medium.

ABSTRACT. The authors will discuss the limitations of traditional postural instruction in typical music school settings, and describe their posture training program for university music students. Recommendations will be offered for those who would like to increase their understanding and ability to implement efficient performance strategies for themselves and their students. A rubric will be provided for the application of the intervention principles. Following this, as time allows presenters will work with interested musicians, tailoring a program for their specific needs. These participants will be screened for biomechanical and neuromechanical dysfunction related to posture. A specific program will be developed to improve static and dynamic posture as it relates to performance.

ABSTRACT. We associate the bass hammer-string contact duration with the time of propagation of compression wave traveling through the hammer body from the contact point to the hammer kernel and back. Based on the hereditary model of the microstructured wool felt, it has been revealed that the stiffness of felt is a nonlinear function of the compression, and it is strongly determined by the rate of the felt loading. This means that the speed of the compression wave that propagates through the felt depends on the wave form and on its amplitude. It has been shown that the pulse of a smooth form, and which has no discontinuity on its front propagates with the constant speed until the accumulation of nonlinear effects results in the eventual continuous wave breaking. After that moment the shock has been formed, and now the velocity of the shock wave depends on the value of the jump discontinuity across the wave front. It has been shown that the front velocity of the shock wave is greater than the velocity in a linear medium. Therefore, the total time of wave propagation, which is related with a duration of the hammer-string contact, decreases as the dynamic level of the hammer impact is raised. As result, for the first bass hammers, the contact duration is shorter than the round-trip time to agraffe, and hence, no reflected wave is needed to assist the hammer for going away from the string.

ABSTRACT. A nonlinear model for a vibrating Timoshenko beam in non-forced unknown rotation is derived from the virtual work principle applied to a system of beam with mass at the end. The system represents a piano hammer shank coupled to a hammer head. An energy-based numerical scheme is then provided, obtained by non classical approaches. A major difficulty for time discretisation comes from the nonlinear behavior of the kinetic energy of the system. This new numerical scheme is then coupled to a global energy-preserving numerical solution for the whole piano. The obtained numerical simulations show that the pianistic touch clearly influences the spectrum of the piano sound of equally loud isolated notes. These differences do not come from a possible shock excitation on the structure, nor from a changing impact point, nor a ``longitudinal rubbing motion'' on the string, since neither of these features are modeled in our study.

ABSTRACT. The present work studies the double polarisations phenomenon observed in vibrating piano strings. When the string is given an initial displacement in one transverse direction (e.g. in the vertical plane), it is found out that the second transverse displacement (e.g. in the horizontal plane) will also be excited and the amplitude can be of similar order to the first transverse displacement. This phenomenon contributes to a characteristic piano sound feature called the "double decay". The understanding of the physical origin of the double polarisations may be attributed to the complex boundary condition observed in piano strings, or to the nonlinearities due to large-amplitude vibrations. The purpose of this study is to clarify their respective roles, by combining modelisation and experiments. The nonlinear vibrations are first studied with a two-degrees-of-freedom system extracted from the Kirchhoff-Carrier string equations. The method of multiple scales is used to study the free vibrations of two polarisations having nearly equal eigenfrequencies and thus presenting a 1:1 internal resonance. It is found out that depending on the energy of the excitation, an uncoupled transverse mode can develop into a coupled mode where there is energy exchange between the two transverse polarisations. The coupled mode is stable and the string oscillates in an elliptic path. Numerical experiments are also carried out, confirming the findings of the analytical approach. Next steps of the research will focus on the complex boundary condition by developing a lumped bridge model. Experiments will be conducted in order to properly discriminate the two effects and quantify their relative importance in the appearance of double polarisations in real piano strings.

ABSTRACT. The piano soundboard is an orthotropic plate made of spruce, ribbed by multiple stiffeners (the ribs) parallel to the grain direction of the main panel’s wood and two addition beams (the bridges) nearly in a perpendicular direction. This complex structure transforms the piano string vibration (coupled to the soundboard at the bridge) into sound; its vibrations and radiations are therefore of primary importance for the sound characteristics of the instrument. Several vibroacoustics models have been developed these last decades using different methodologies: finite element / numerical modeling (Berthaut et al., Applied Acoustics, 2003 ; Chaigne et al., JASA, 2013 ; Chabassier et al., JASA, 2013) or reduced models using global descriptors (Boutillon and Ege, JSV, 2013). An analytical model recently developed at LVA (Trévisan et al., ISMA 2014 / NOVEM 2015) is particularly well-adapted for a parametrical study and appears as an alternative to time-consuming numerical methodologies. The model is based on a variational approach that takes into account plate and superstructures energies. The soundboard vibration is decomposed on the corresponding orthotropic simply supported unribbed plate modes. The aim of this analytical tool is ultimately to help piano manufacturers to predict the influence of structural modifications of the soundboard (number/dimensions of ribs/bridges…) on the sound of the instrument. In order to validate the methodology and hypotheses done in the analytical model, we present and compare in this communication measured/predicted vibroacoustics quantities obtained for a same structure and under same (supposed) conditions: a Pleyel P131 upright piano soundboard fixed on its wooden rim. The quantities compared are modal basis in the low-frequency domain [0-400Hz] and point mobility along the bridge for a larger frequency band [0-5kHz]. Results on vibrations are very satisfying demonstrating the validity of the model. Experimental radiation results (soundboard radiated power) will also be given and compared to predicted quantities.

ABSTRACT. The leading idea of this theoretical paper is to examine the effects of structural changes in the piano on the basis of energetic quantities relative to its constitutive parts. These energies are global quantities which characterize the intrinsic properties of the instrument, irrespective of the observation point. The evolution of the various energy terms with time are calculated with a help of a recent piano model which couples together the hammer, the nonlinear strings, the soundboard and the acoustic space. Some parameters, which play a major role in the history of the piano are particularly examined: string tension and diameter, soundboard thickness and rigidity, hammer mass and velocity. The results show that direct links can be established between the energetic quantities and the tonal properties of the piano sounds, in terms of temporal envelope and spectral content. they also shed useful light on the energy exchange between the constitutive parts of the instrument, and on its acoustical efficiency. This study is intended to have potential applications as a theoretical guideline for piano making, restoration and reproduction of historic instruments.

ABSTRACT. Results from three experiments are presented, showing that the perceived acoustic and vibrotactile quality of a reproduced piano does not require models simulating every aspect of the original instrument with great accuracy. It was found that high-quality loudspeaker array passive listening at the pianist's position admits distortions of the sound field. Furthermore, pianists during playing seem to compensate for errors in the auditory scene description. Finally, they are particularly sensitive to the existence of vibrotactile feedback on their fingers meanwhile tolerant about the precision with which this feedback is reproduced. Based on these results we are currently working on a lightweight portable physics-based digital piano design, that should improve upon the experience a pianist with no keyboards at hand makes when interacting with a touch-screen piano software running on smartphones and laptops.