A Experimental Study of a Laminar Separation Point.

Measurements in the neighborhood of a laminar separation point at a high Reynolds number are presented, together with a discussion of the validity of proposed theories for such a point. Most analyses of laminar separation at high Reynolds numbers make use of the boundary-layer equations in the neighborhood of separation. It is well established that solutions to these equations for a prescribed external velocity exhibit a singularity at the separation point, which is unphysical. This may be caused by inaccuracy of the boundary layer equations, neglect of the interaction between the boundary layer and the external flow, or some other factor which has not yet been identified. Inverse solutions of the boundary layer equations, in which the external velocity is not used as a boundary condition, do not exhibit singular behavior at separation. This suggests that the boundary layer equations might be valid at separation. The purpose of the experimental program was to expose the physics of the neighborhood of a laminar separation point, in order to shed some light on this situation. A circular fitted with a long splitter plate was subjected to a uniform flow, such that the Reynolds number, based on the cylinder's diameter, was 4.68 x 10('4). The velocity field was measured using a dual-beam laser-Doppler anemometer with optical frequency shifting. The wall shear stress was measured by a hot-film sensor: the directional ambiguity was corrected by comparison with the LDA data. The wall pressure was measured using a strain-gage pressure transducer. It was found that there was no evidence of a singularity at the separation point; instead, the behavior predicted by the Navier-Stokes equations was observed. A number of calculations, amongst which was the calculation of individual terms in the Navier-Stokes equations from the experimental data, led to the conclusion that the boundary layer equations are valid at, and near, a laminar separation point.