A methodology based on harmonic response methods and vibroacoustic analysis was developed by the authors in order to compute and predict the acoustic field emitted by an external gear pump. A quite simplified FEM model of a 8.5 cc/rev external gear pump was developed with fixed support between the flange and the test rig and the pump was considered as a "black box". A first comparison between computed and experimentally measured noise levels at the main modal frequencies for some pump working conditions allowed us to improve the pump FEM model. This paper reports some results of the validation process of this methodology. In particular, the results of sound power tests performed on the pump at some working conditions were compared with the predicted sound power values calculated by applying this numerical method. The sound power tests, based on sound intensity measurements, were carried out on a test rig complying with the requirements of ISO 16902 standard. The pump was completely isolated from the prime mover and was mounted on a reflecting plane. A parallelepiped, centred with respect to the pump, was chosen as measurement surface. Sound intensity measurements were performed by scanning procedure according to ISO 9614-2 test code at three different working conditions. In parallel, for each working condition the sound intensity field emitted by the pump was numerically computed in a grid of points on the same parallelepiped surface. Results confirmed the potentiality of this approach in offering a good compromise between noise prediction accuracy and reduction of experimental and modelling requirements.
External gear pump noise field prediction: Sound intensity analysis and methodology validation
Miccoli Giuseppe;Pedrielli Francesca;Carletti Eleonora;Parise Giorgio
2016
Abstract
A methodology based on harmonic response methods and vibroacoustic analysis was developed by the authors in order to compute and predict the acoustic field emitted by an external gear pump. A quite simplified FEM model of a 8.5 cc/rev external gear pump was developed with fixed support between the flange and the test rig and the pump was considered as a "black box". A first comparison between computed and experimentally measured noise levels at the main modal frequencies for some pump working conditions allowed us to improve the pump FEM model. This paper reports some results of the validation process of this methodology. In particular, the results of sound power tests performed on the pump at some working conditions were compared with the predicted sound power values calculated by applying this numerical method. The sound power tests, based on sound intensity measurements, were carried out on a test rig complying with the requirements of ISO 16902 standard. The pump was completely isolated from the prime mover and was mounted on a reflecting plane. A parallelepiped, centred with respect to the pump, was chosen as measurement surface. Sound intensity measurements were performed by scanning procedure according to ISO 9614-2 test code at three different working conditions. In parallel, for each working condition the sound intensity field emitted by the pump was numerically computed in a grid of points on the same parallelepiped surface. Results confirmed the potentiality of this approach in offering a good compromise between noise prediction accuracy and reduction of experimental and modelling requirements.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.