Elastodynamic analysis of vibratory bowl feeders: modeling and experimental validation

Vibratory bowl feeders are often used for conveying and feeding small spare parts in automatic assembly systems. They consist of a bowl connected to a base by three or four inclined leaf springs. The springs constrain the bowl so that its vertical displacement causes a coupled rotation around its vertical symmetry axis. One or more electromagnets generate the sinusoidal force which drives the bowl; commonly, they are either vertically or tangentially housed between the base and the bowl. This force induces the movement of the spare parts along the inner spiral track of the bowl. Rubber mounts are located under the base in order to isolate the feeder vibration from the floor. The feeding velocity of the parts is influenced by the vibration amplitude of the bowl. Most vibratory feeders are used at the resonant or near-resonant frequency of the mechanical system in order to improve feeding efficiency. Furthermore, direction, amplitude and frequency of the oscillations depend on the design parameters of the feeder. In the literature, a number of theoretical and/or experimental studies have been reported for this device (see for instance [1,2]). Moreover, other works have addressed the analysis of the spare parts’ motion through numerical models (see, for example, [3,4]). Beside a simple construction, the main disadvantages of these feeders are the noise generation as well as the induced vibrations on the surroundings during operation. In this sight, the authors have developed a general elastodynamic model of the feeder for predicting its dynamic behavior. The model takes into account the most important parameters involved during operation: leaf spring stiffness, base and bowl inertia, forces due to the electromagnets, damping effects and dynamic stiffness of the base mounts. All these parameters have been also experimentally evaluated on a real feeder used in automatic assembly lines by pharmaceutical companies. Globally, the proposed model is a lumped-parameter model with three degrees of freedom. The proposed model has been validated by using two methodologies: the experimental modal analysis, and the operational vibration measurements. The results of these validations show that the model can be used both to analyze the dynamic behavior of the feeder, and to evaluate the effectiveness of parameter changes during design