Reproducibility of modal parameters using different excitation types

Experimental acquisition techniques of modal parameters are well-known topics in the dynamic characterization process of steel structures. The objective is to determine the structure's natural vibration properties, which are eigenfrequencies, corresponding mode shapes and damping. Mostly, the experimentally obtained modal parameters are used to verify numerical models, which can be utilized e.g. for overall system vibration simulations. The modal parameters were taken from single measurements and thus, few information about the parameter variances are available. In experimental modal analysis, the choice of the boundary condition (e.g. free or grounded support) as well as the corresponding excitation type (e.g. impulse hammer or shaker) may impact the variance of the obtained modal data. The present article focuses on the difference in reproducibility of modal parameters using an impact hammer and a modal shaker, where measurements are carried out under free/free boundary conditions. The test object is a gear wheel, which first eigenfrequencies occur in a frequency range around 4.6 kHz. During the experimental modal analysis, the frequency response functions are obtained up to 20 kHz. Due to the high frequency range, a 3D scanning laser Doppler vibrometer is used for system response acquisition for all investigations. In a first test setup, statistical variations of a single gear wheel's modal parameters are analyzed and special attention will be payed on the reproducibility of the excitation force using an automated impulse hammer. In a second step, the differences of modal parameters of three gear wheels from one lot are presented. Finally, repeating measurements are carried out using a modal shaker for system excitation and the corresponding modal parameters are compared to the results obtained using an impact hammer.