| Budget Amount *help |
¥5,500,000 (Direct Cost: ¥5,500,000)
Fiscal Year 1988: ¥3,100,000 (Direct Cost: ¥3,100,000)
Fiscal Year 1987: ¥2,400,000 (Direct Cost: ¥2,400,000)
|
|---|
| Research Abstract |
Turbulent swirling flows along annulus formed between two concentric stationary cylinders have been studied. And the summary of the results is shown below.
1. EXPERIMENTAL RESULTS
Firstly, inner and outer wall pressure was measured under the seven different swirl angles changes by I.S.V.(inlet guide vanes), and the thrdd different reynolds number(1*10^5,2*10^5,3*10^5). The measured flows were divided into three groups by the weirl angles, and the critical intensity of swirl was consistent with it of the swirling flows along the stationary cylinder studied by murakami et al.. Secondly, three kinds of swirl. angles( =0',30',50') were selected from these three groups, and traverse stations were located three different axial distance(Z/DH=0.0,6.85,12.3). Mean velocity, turbulent intensity and pressure profiles from viscous sublayer to outer layer were measured at three stations with three swirl angles. By measurements, these following results were obtained;
With =0h, mean velocity profiles ne
… More
ar inner and outer wall showed the good agreement with the logarithmic law of the 2-D flow along a flat plate. But with =50', the gradient of the absolute, axial velocity near inner wall was larger and near outer wall was less than 2-D flow. And with =30', the effect of swirl( due to the centrifugal force ) was less remarkably than =50', logarithmic law of the absolute, axial velocity were in keeping agreement with 2-D flow, and logarithmic law of the circumferential velocity was not different between near inner wall and near outer wall. With every swirl angles, the boundary layer near inner and outer wall was not skewed from viscous sublayer to buffer layer.
2. NUMERICAL SOLUTION
By using isotropic and anisotropic k- model, the flows were calculated. And the weak points of k- model became clear as follows;
When the two-components of eddy viscosity ^ rz, ^ re were equal to the scalar eddy viscosity ^ t, the calculation results with =50' were no more than meraly poor agreements with the measurements. And with =50', the model constant of eddy viscosity, C_ =0.09, were inconsistent with the measurements. Less
|
