Tonalness and consonance of technical sounds

• Tonhaltigkeit und Konsonanz von technischen Geräuschen

Fingerhuth, Sebastián; Vorländer, Michael (Thesis advisor)

Aachen : Logos Verl. (2010)
Dissertation / PhD Thesis

In: Aachener Beiträge zur technischen Akustik 10
Page(s)/Article-Nr.: X, 138 S. : Ill., graph. Darst.

Aachen, Techn. Hochsch., Diss., 2009

Abstract

In this work, the production and perception of noise from technical equipment is presented. As a case study, the noise of switched reluctance machines (SRM) was used.The first chapter of this dissertation is the introduction and refers to the noise problem, specially of technical origin. Therefore several psychoacoustic magnitudes and parameters that are widely used for describing sounds are presented, followed by some ideas about noise perception. The concept and applications of sound quality of a product are shown as well as the importance of the information that a sound (or noise) can give to the user of that product (e.g. a machine). Then, some theories and measurement methods on human perception and psychophysics are introduced. Finally the concepts of tonalness and consonance of sound are briefly presented. In the second chapter, the principle of operation of SRM is presented and its torque production is derived. Chapter 3 presents some noise problems of electrical machines. The vibroacoustic consequences of the radial forces acting on the stator teeth is a typical SRM problem. In the fourth chapter the vibration characteristic of the machine is described. A simple analytic model (using two quadrupoles) serves as a first approximation. Afterwards the results of the numerical simulation (FEM-BEM) are presented and compared with the measured data. The constructed eddy-current test bench as well as the multi-channel measurement equipment is presented. The results from the motor housing vibration shows that the vibration behavior of the machine can be described by two mode 2 vibrations. The directivity of the radiated sound from the machine is also presented. In the fifth chapter, the results from a noise optimized control strategy for a SRM are shown, with an example of what is achievable and which are the limits of such strategy. The concepts of tonalness and consonance and some theories behind them are presented in chapter 6. Then, the results of the psychoacoustic listening tests are presented. The listening tests were performed to obtain quantitative information about the perceptual characteristics of the noise. Each listening test is presented individually, as well as the statistic evaluation and analysis of it. The stimuli used were recordings of machine noise or synthesized sounds, based on machine recordings. The main conclusions of this work are:- For the measured 8/6 SRM, the most important vibration problem is a cylindrical mode 2 vibration. The eigenfrequency of this mode is in the frequency range where the hearing system is most sensitive and the radial force excites mainly this mode.- One important component of the perceived annoyance of technical sounds is the tonalness.- The comparison of different tonalness calculation algorithms showed that Aures’ had a high agreement with the results from the listening tests.- The consonance/dissonance of technical sounds depends on several parameters. Roughness is probably the most important, which is highly dependent on the frequency ratio between tonal components.- Using two different evaluation methods for the listening tests (magnitude estimation and category partition) gave the same results for the consonance listening test.