Abstract:
In order to obtain a reasonable prediction of the flow through the vocal folds, for applications such as a two-mass model for voiced sound production, the use of a quasistationary flow model is considered. This model assumes a quasisteady jet and the flow separation is predicted by means of a boundary layer model. Limitations on the validity of such a model are investigated by means of in vitro experiments which provide global information in the form of flow visualizations and quantitative information in the form of wall pressure and hot wire measurements. A two-dimensional numerical simulation based on a viscous blob method provides further quantitative information on the details of the flow. However, such a simulation cannot predict the transition from laminar to turbulent jet flow which is an essentially three-dimensional effect. Furthermore, both the simplified model and the numerical simulation do not predict the occurrence of asymmetry in the flow which is sometimes observed in the experiments. Imbedding the numerical calculation of the locally incompressible flow in an acoustical model will be discussed. [Work supported by Technology Foundation Grant No. ETN33.2952.]