Detailed measurements of pressure distributions, mean velocity profiles and Reynolds stresses were made in the thick, axisymmetric boundary layer and the near wake of a low-drag body of revolution. The data are presented in graphical as well as tabular form for convenience in later analysis. These measurements shed some light on the joint influence of transverse and longitudinal surface curvatures and pressure gradients on the boundary-layer development and on the manner in which an axisymmetric boundary layer becomes a fully-developed wake. Apart from giving a complete set of data on such an important flow configuration, the measurements should provide a fairly rigorous test case for some of the recent turbulence closure models which claim a level of generality not achieved by the older phenomenological models. By inclusion of recently proposed modifications to account for the effects of the extra rates of strain on the turbulence length scale arising from longitudinal and transverse surface curvatures, it is shown that the boundary layer in the tail region of a body of revolution is dominated by the extra strain rates and that more research is needed to account for them properly even in the most recent calculation procedures.

Detailed measurements of pressure distributions, mean velocity profiles and Reynolds stresses were made in the thick, axisymmetric boundary layer and the near wake of a low-drag body of revolution. These measurements shed some light on the joint influence of transverse and longitudinal surface curvatures and pressure gradients on the boundary-layer development and on the manner in which an axisymmetric boundary layer becomes a fully-developed wake. The present data have been used to provide an independent check on the accuracy of the simple integral method proposed by Patel, and its extension to the calculation of the near wake made by Nakayama, Patel and Landweber. Calculations have also been performed using the differential equations of the thick axisymmetric turbulent boundary layer and a rate equation for the Reynolds stress derived from the turbulent kinetic-energy equation along the lines suggested by Bradshaw and others. It is shown that the boundary layer in the tail region of a body of revolution is dominated by the extra strain rates arising from longitudinal and transverse surface curvatures. A new differential method is incorporated into the iterative procedure developed by Nakayama, Patel and Landweber for the solution of the interaction between the boundary layer, the wake and the external inviscid flow. The results of the iterative method have been compared with the experimental data obtained from the present low-drag body and those obtained earlier on a modified spheroid to demonstrate agreement. (Author).

An examination of the Reynolds equations for axisymmetric turbulent flow where the thickness of the boundary layer is of the same order as the transverse radius of curvature of the surface shows that neither the boundary layer nor the potential flow outside it may be calculated independently of the other, owing to significant interactions between the two flow regimes. Following a discussion of various procedures for extending conventional thin boundary-layer calculation methods to treat thick axisymmetric turbulent boundary-layers, a method is proposed for the simultaneous solution of the boundary layer and the potential flow equations, allowing the two flow regimes to interact. (Author Modified Abstract).

An analysis is presented which predicts the properties of an arbitrarily thick turbulent boundary layer in axial flow past a long cylinder. The study makes use of a modified form of the turbulent law-of-the-wall, which properly accounts for transverse curvature effects. Using this law, the theory which follows is then an exact solution to the axisymmetric equations of continuity and momentum in incompressible flow. Numerical results are given to show the effect of curvature on the various boundary layer characteristics. Skin friction and drag coefficients can be increased greatly with increasing curvature while boundary layer thickness is decreased. When defined in their axisymmetric form, the displacement and momentum thickness are both decreased by curvature. The velocity profile is flattened greatly and the shape factor approaches unity at large curvature. The failure of earlier power-law theories to make accurate predictions is shown to be due to their inadequate handling of the strong profile shape changes. (Author).