An analysis is presented of a plane turbulent far wake of an incompressible fluid with a modified expression for eddy viscosity. Considering the kinematic eddy viscosity is expressed as the time average of the product of the mixing length and the lateral fluctuation velocity, it is assumed that the latter varies laterally like its rms value, which has been measured, and is approximated by a power of the mean velocity defect. Suitable values of the exponent are inferred by applying the TOWNSEND's data. The semi-empirical formula obtained finally for the mean velocity defect is similar to the formula obtained by SCHLICHTING, but is free from any singularity, and is found to express the available experimental measurements rather well with some variation of the value of the exponent to each measurement.
The density gradient is, in some cases, observed near the body surface when a pair of vortices in vortex wake, which develops behind vehicle flying in supersonic speed with incidences, is still stable and their strength is enough strong. The purpose of this paper is to examine the mechanism responsible for the generation of this density gradient and to find out it's characteristics. The systematic experiments including pitot tube traverse and schlieren photographing are conducted and further the computational simulation of the flow field are carried out, the flow model of which is two dimensional jet impingement. Their results are readjusted as follows. 1. There is an entropy jump across this density gradient. 2. The flow field caused by interaction between the boundary layer and this density gradient turning down by other shock wave is similar to that of shock-boundary layer interaction. 3. The position of shock wave obtained by computational simulation is almost coincident with that of this density gradient observed on the schlieren photographs. On the base of the reasons above mentioned, it is confirmed that this density gradient is shock wave, which is produced by blowdown of a pair of vortices. And if there are stable vortices being enough strong and their space is favorable, it is concluded that this shock wave is surely observed near the body surface in vortex wake even in the cases of wider range of Mach number and more complicated body shapes.
This paper presents an approximate solution of the second-order boundary layer equations for a two-dimensional laminar wall jet of an incom- pressible fluid over a curved surface. Applying the small perturbation method with a small parameter expressed as a power of the streamwise coordinate, similar and non-similar velocity pro- files are calculated along curved walls when the local curvature is proportional to xm-3/4, in which m=0 corresponds to the hitherto treated case of the similar flow, m=3/4 to the flow over a circular cylinder and m>1 to the flow accelerated downstream in the case of convex surfaces. The variation of the velocity profile, the total entrainment rate, the skin friction and the surface pressure as a function of curvature or the exponent m are indicated.
As a sequel to the previous paper (Ref. 2), numerical methods are investigated for calculating the pressure distributions on an airfoil with oscillating flap in incompressible flow and on an oscillating airfoil in compressible flow. The logarithmic singurarities in kernel functions are manipulated in the same way as our previous paper. Accuracy of numerical solution in compressible flow is estimated introducing an appropriate error-evaluation index which compares normalwashs, since there exist no exact solutions. Throughout this paper, some existing suitable interporation functions are utilized.
Finite element analysis of the experimental data of YEW and RICHARDSON on torsional plastic waves in a tube of copper is presented. Three different constitutive equations are used to describe the strain data in torsional impact. The strain-rate dependent solutions show better agreement with the experiment than the strainrate independent solution, but experimental observations that the plastic shear strains along the tube decrease with increasing distance from the impact end are not well predicted by any one of the rate-dependent, or rate-independent theories considered. The discrepancies between experimental and computed results are due to the inaccuracy of constitutive equations employed in the present study. It appears therefore that the more realistic constitutive equation must include not only strain-rate effect but also some factors such as strain-rate history and temperature effects.
An experimental investigation of the unsteady force acting on a cascade blade due to an upstream moving cylinder has been carried out in order to compare with the computed results presented in the 1st report. The prediction agreed well with the experimental one excepting a few discrepancies. The reason why these discrepancies arise was considered.