In this paper the vibro-acoustical properties of the EMFi-actuator are studied. The examination is carried out by modelling the operation of the actuator and by comparing the modelling results to the corresponding measured values in case they are available. The modelling is based on an analytical approach. The equations used are derived in the Appendices. Consideration on the properties of the vibration is divided into three parts: the basic vibration of the film, vibration of the film in the time domain and vibration of the functional elements of the actuator. The sound production properties of the actuator have been examined in the free acoustic field and in cases with boundaries near the source. The effect of the different vibration distributions of the actuator film and impedance loading for the emitted sound has been studied. Also the radiation pattern of the actuator is modelled and compared to the measured results. The general vibro-acoustical functioning of the EMFi-actuator is stated in the impedance-oriented form by evaluating the loading impedance of the vibrating film from its boundary conditions. The basic vibration functioning of the actuator film in the frequency domain is predicted by using a linear second-order ordinary differential equation with constant coefficients. In the time domain the non-linear effects of the spring force and electric forces acting on the actuator film are considered. The recoil effect of the actuator is studied with a normal-mode method of dynamic analysis. It is found that the response of the supporting structure may decrease with about a decade without influencing the response of the actuator film. The use of the arithmetic average value of the vibration deflection of the EMFi-actuator in the prediction of sound power or sound pressure in the acoustic far field at low frequencies is compared to the frequency or spatially dependent vibration of the actuator film. The value of the arithmetic average of vibration is valid even when the outward impedance of the film is not taken into consideration. The large dipole type actuators are more effective sound radiators than monopole sources. The effect of the thickness of the absorbent material under the dipole actuator, the radiation pattern and the effect of the boundary conditions of the vibrating cell on the radiating sound power are considered. The measured and predicted values of radiation patterns are very similar in every angle. The rigidly supported boundary of the actuator cell enables the actuator panel to radiate much more sound power with all the different even modes than it is possible with a simply supported boundary condition. The basic problems concerning the operation of an EMFi-actuator are related to its ineffective low frequency sound production and the distortion caused by non-linear vibration. It is possible to increase the sound power radiation properties of the monopole or dipole type functional modes of the actuator by making the deflection of the membrane larger (using smaller tension), by using larger membrane areas or by using two interacting actuators. The dimensions of the actuator element cell affect the power the single vibration mode is able to radiate into the acoustic far field. When the film cells of the actuator panels are simply supported, the larger dimensions compared to those of the present actuator type, produce more sound power at odd and even modes. The confidence intervals of a normal distribution used in the evaluation of the reliability of the sound power measurements have been obtained by calculating combined standard deviation values from standard deviation values of single frequency bands given in the standards. It is shown that it is possible to find an upper limit for the confidence interval. The main results of this work are the means to increase the radiated sound power of the actuator element without increasing concurrently the distortion of the radiator. These resources are based on the functioning mode of the actuator (monopole, dipole), boundary conditions of the actuator element or cells of the actuator element, the larger actuator areas, interaction of actuators and the dimensions of the actuator element cells. Another result of this study is the potentiality to evaluate the A-weighted standard deviation of the sound power of the equipment to be measured from the measurement results without specifying the type of emitted sound radiated by the equipment.
|Place of Publication||Espoo|
|Publication status||Published - 1999|
|MoE publication type||G3 Licentiate thesis|