The goal of the present study is to provide a simple method to compare structure borne noise sources in order to choose the most efficient one, considering the transmission of dynamic forces. It is well known that mechanical sources are not only dependent of the source itself but also of the receiving structure, in addition real sources cannot be reduced to a transverse force acting on the structure but more complicated effect like moment excitation must be taken into account.The advantage of the reception plate method is to characterize the source globally by the level of vibration of the reception plate whatever the type of excitation, the idea is basically to characterize mechanical sources as it is done for acoustical sources in reverberant rooms.A reception plate test bench has been developed to determine the power injected by mechanical sources. Two prototype plates have been designed in order to have different receiving mobilities. The reception plates have been characterized through their damping factors, modal overlaps, and diffusivity of the vibration fields measured by Laser Doppler Vibrometer (LDV). The goal is to perform a statistic analysis to evaluate the influence of number and position of sensors on the averaged velocity values over the reception plate on the accuracy of the measurements. This quantity allows to estimate the injected power into the receiving structure.A shaker (broadband excitation) and a small pump used in automotive industry (tonal excitation) have been characterized by the test bench. Comparison of injected power allows us to rank source efficiencies.