THE JOURNAL OF THE ACOUSTICAL SOCIETY OF JAPAN
Online ISSN : 2432-2040
Print ISSN : 0369-4232
The Vibration Analysis of A Cone Loudspeaker by the Finite Element Method
Takafumi UenoKen'ichi TakahashiKiyoshi IchidaShin'ichiro Ishii
Author information
JOURNAL FREE ACCESS

1978 Volume 34 Issue 8 Pages 470-477

Details
Abstract
This paper concerned with the vibration analysis of a cone loudspeaker by the finite element method (FEM). The factors which influence the frequency response of the sound pressure level (SPL) of a loudspeaker are disccused in detail. We used NASTRAN as a program for FEM. Our automatic data generator is employed in order to simplify the engineers' task of input data preparation. The frequency response of SPL is obtained by integrating the velocity of each node. A storage graphic display is used to show the shapes of the characteristics of loudspeakers. The calculation of SPL requires about 18 minutes at CPU time with an IBM 370/148 computer. We consider the influences of the cone, the surround (outer suspension), the spider(inner suspension) and the voice-coil bobbin on SPL of a cone loudspeaker. In this caalculation the parts are divided into axisymmetric conical shell elements. The calculation error of the maximum peak frequency of SPL (f_h) is 6% in Fig. 4. First, we considered the influences of the shapes of cones. The first-order eigenvalue (f_1) of a 0. 17mm thick conical cone with a semi-apex angle of 68 degrees is three times that of a 0. 17mm thick curved cone with a curviture of 0. 046m. The thickness dependence of f_1 is shown in Fig. 6. f of the curved cone varies with the thickness of the cone. However, f_1 of a conical cone with a small semi-apex angle is hardly affected by the thickness of the cone. This implies that f_1 of the courved cone depends on bending vibration. Second, we discussed the influences of the surround. The peaks and dips of SPL from the surround are caused by natural resonances. The maximum peak frequency of the tested cone suspension is about 5 kHz. As is shown in Fig. 10, this resonant frequency is inversely proportional to the radius of the curviture of the surround and proportional to the sound velocity in its material. Third, the influences of the voice-coil bobbin and the inner edge of a cone are discussed. The thickness, elasticity and length of the voice-coil bobbin affects the frequency range of SPL. Finaly, the resonance of the spider is shown in Fig. 13. The first-order resonant frequency of the tested spider is shown in Fig. 13, and about 777Hz. The measured vibration pattern fairly agrees with the calculated one. This resonance of the spider decreases the smoothness of the frequency response of SPL.
Content from these authors
© 1978 Acoustical Society of Japan
Previous article Next article
feedback
Top