| The purpose of this study was to experimentally elucidate the key physics of Coanda ejectors and to find a model that could both predict the performance of Coanda ejectors and aid in their design.;Velocity profiles, turbulence intensities, turbulence spectrums, correlation functions and wall pressure distributions were measured inside two different Coanda ejectors at different operating conditions.;Using laser anemometry, axial velocity profile measurements were done inside both ejectors. This was accomplished by angling the laser beams with respect to the axis of the ejectors. All velocity profiles measured had similar characteristics--a maximum near the ejector's wall decreasing to a uniform and constant value in the center of the ejector.;Based on the experimental results and our analysis of these results, we have postulated a mechanism for the operation of the Coanda ejector: The driving force for the ejector's operation is the primary flow momentum. The turbulent mixing of the primary flow with the ambient air near the entrance of the ejector, transfers this momentum to the stagnant air and induces a secondary flow. This secondary flow is then dragged by turbulent shear forces downstream to the ejector exit while being mixed with the primary flow by the persistence of a large turbulent intensity throughout the ejector.;With the key flow properties of the ejector elucidated by these experiments, we have examined several theoretical models to find a model that could predict the performance of these ejectors.;Normally two approaches have been used: Control Volume Approach (CVA) and Physical Phenomena Approach (PPA). The results have shown that those models can predict the gross behavior of Coanda ejector. However, neither can predict the detailed performance of Coanda ejectors nor can they be used for design purposes. Due to this deficiency we have developed a new model.;This model is based on the experimental results. Using this model, we have predicted the average velocities at any section of the Coanda ejector with the maximum difference of 9% between the predictions and measurements. Also, we have predicted the ratio of the Induced Flow Rate/Primary Flow Rate (I/P) with a maximum difference of 5%. (Abstract shortened by UMI.)... |
