This session aims to provide a platform for all aspects of computer analysis of musical instruments. This includes analytical simulation methods, virtual sound generation, numerical computation algorithms, computer optimisation, acoustical geometry and other parameter reconstruction and especially their application to practical problems.
|Curtit; Marthe: |
(Invited) / O
|'BORE RECONSTRUCTION BASED ON INPUT IMPEDANCE MEASUREMENT'|
|The bore profile of a wind instrument has an important influence on its acoustical characteristics. Specifically, playing frequencies are close to the resonance frequencies of the tube constituted by the bore of an instrument. Several methods were developed to obtain bore dimensions, such as pulse-reflectometry, or radiography techniques. The method investigated here is based on tube input impedance measurement, witch provides us with an alternative way to estimate the bore profile. |
In collaboration with Olivier Cottet, we have applied this method in the case of a bassoon crook, at different stages of fabrication. During the crook fabrication, a sheet of metal is used to form a straight tube which internal profile is given by a mandrel of which dimensions can be measured precisely. This tube is then filled with alloy lead and curved. At this stage, taking mechanical measurements of the bore (using a calliper for example) is not possible due to its small size. The aforementioned method is then used to make those measurements, and maybe to observe some differences in the bore profile, after and before the crooks is curved.
|Sharp; David: |
(Invited) / O
|'INVESTIGATING OBOE MANUFACTURING CONSISTENCY BY COMPARING THE ACOUSTICAL PROPERTIES OF FIVE NOMINALLY IDENTICAL INSTRUMENTS '|
|For large-scale musical instrument makers, the ability to produce instruments with exactly the same playing characteristics is a constant aim. Modern acoustical measurement techniques (such as acoustic pulse reflectometry and input impedance measurement methods) together with psychoacoustical testing, can help this goal be reached. This paper investigates the issue of instrument manufacturing consistency by comparing the acoustical properties and the perceptual qualities of five Howarth S10 student oboes. Input impedance measurements have been made on the five oboes for fingerings throughout the standard playing range of the instrument, acoustic pulse reflectometry has been used to measure the bore profiles of the oboes, and nine musicians have taken part in a two-alternative-forced-choice discrimination playing test using two of the instruments. The main findings are (i) the instruments are perceived as identical by most of the musicians tested, (ii) a variation in the playability of the note F6 experienced by two of the musicians is shown to be due to differences in the elevation of the pad above the C hole, and (iii) some small variations in the playing properties in the first register of the instruments are shown to be related to differences in input impedance which, in turn, appear to arise from small differences in the bore profiles of the instruments.|
|Chatziioannou; Vasileios: / P||'PARAMETER OPTIMISATION FOR WOODWIND SINGLE-REED MODELS. (V. CHATZIIOANNOU, M. VAN WALSTIJN)'|
|Time-domain modelling of single-reed woodwind instruments usually|
involves a lumped model of the excitation mechanism. The parameters of
this lumped model have to be estimated for use in numerical simulations.
Several attempts have been made to estimate these parameters, including
observations of the mechanics of isolated reeds, measurements under
artificial or real playing conditions and estimations based on numerical
In this study an optimisation routine is presented, that can estimate
reed-model parameters, given the pressure and flow signals in the
mouthpiece. The method is validated, tested on a series of numerically
synthesised data. In order to incorporate the actions of the player in
the parameter estimation process, the optimisation routine has to be
applied to signals obtained under real playing conditions. The estimated
parameters can then be used to resynthesise the pressure and flow
signals in the mouthpiece. In the case of measured data, as opposed to
numerically synthesised data, special care needs to be taken while
modelling the bore of the instrument. In fact, a careful study of
various experimental datasets revealed that for resynthesis to work, the
bore termination impedance should be known very precisely from theory.
An example is given, where the above requirement is satisfied, and the
resynthesised signals closely match the original signals generated by
|Chick; John: / O||'THE EFFECT OF MOUTHPIECE DESIGN ON SLURRED TRANSIENTS IN BRASS INSTRUMENTS'|
|For a musician the choice of a specific instrument is almost always some sort of compromise. The player will be assessing the timbral qualities, intonation and response in different registers and at different dynamic levels, and the ease with which the player can slur between notes is normally one of the key response characteristics used in this assessment. The mouthpiece is known to be highly significant in determining how the instrument plays but it is not clear what its effects are on slurring between notes. The research presented here uses a combination of experimental data from player tests, and recently developed time domain models, to investigate the effect of mouthpiece design parameters on slurred transients.|
|Kemp; Jonathan: / O||'RADIATION AND REFLECTION IN A RECTANGULAR CROSS-SECTION WAVEGUIDE: FINITE DIFFERENCE SIMULATION AND THEORETICAL FREQUENCY DOMAIN IMPEDANCE.'|
|If a theoretical expression is known for the radiation impedance then it may be projected to predict the input impedance and input impulse response in an acoustic waveguide. Radiation impedance may be derived from integration of Greenís functions and so are based on continuous expressions in the frequency domain. |
In this study the finite difference technique will be used to simulate the reflections of a band limited impulse as it travels down a horn and partially radiates and reflects from the end. The finite difference technique used here is based on approximating the 3 dimensional lossless wave equation using differences between adjacent grid points to calculate differentials. These calculations will be performed for a discretized 3D rectangular horn within a large discretized 3D box whose walls may be set to absorb incident energy.
The finite difference results will be used to create animations to visualise the pressure field at the point of partial reflection and transmission of a band limited impulse in the time domain. Frequency domain analysis will then be used to compare the results with those predicted by the frequency domain impedance techniques.
|Nederveen; Cornelis: / O||'EFFECT OF TRANSVERSE ACOUSTIC FLOW ON THE INPUT IMPEDANCE OF RAPIDLY FLARING HORNS'|
|For most wind instruments, the waves in the inside of the pipe are approximately plane. The field may be considered one-dimensional; the input impedance can be calculated by describing the pipe as a succession of short cones (transmission line method). This is less accurate for fast changes in the bore, for example in a rapidly flaring horn, where the kinetic energy of the transverse acoustic flow is no longer small compared with that of the longitudinal flow, causing a local increase in inertia, resulting in an effective increase in length when the horn is located near a pressure node. For higher frequencies, where cross-dimensions become comparable to the wavelength, resonances can occur in the cross-direction. To investigate this, the pipe radiating in open space was modelled with a finite difference method. Because of limits in computer capacity the outer domain is bounded. To avoid reflections its boundaries must be fully absorbing just as the walls of an anechoic chamber. The well-known Sommerfeld radiation condition is only useful at low frequencies, reflections occur at higher frequencies. Much more effective is Berengerís PML (perfectly matched layer); it is applied here. Presented are results for various horns, they are compared with earlier published investigations on flanges. In all cases the inertance exhibits a maximum when the largest cross-dimension (outer diameter of flange or diameter of the horn end) becomes comparable to half a wavelength. This effect changes the position of higher modes in the pipe, influencing the conditions for mode locking, important for ease of playing, dynamic range and sound quality.|