Turbulent Decay in Active-Grid Generated Isotropic and Homogeneous Flow with Steel Mesh

To study the mixing of passive scalars in turbulent flow, the challenge has become to generate turbulence that is both isotropic and homogeneous at high intensities. Most experiments have utilized passive grids that generate 2-3% turbulence intensity (TI = u'/ <u> where u' and <u> are the rms and mean downstream velocities, respectively). Previous studies, such as that of Mydlarski and Warhaft, use an active grid but no turbulence manipulator (Mydlarski and Warhaft 1995). Our study uses an active grid, similar to that used by Makita, to create up to 15% turbulence intensity (Makita et. al. 1987). To homogenize the flow, we use fine-spaced, cross-wired steel meshes placed downstream of the active grid. We measure the mean and rms velocity at incremental locations within the tunnel. For different Taylor microscale Reynolds values (R_l = u'λ/ν, where λ is the microscale length, and ν kinematic viscosity), we make comparisons between actual and expected kurtosis and skewness values (S(u) = <u³> / <u² >^3/2 , u being the instantaneous velocity in the x-direction downstream) (Mohamed and LaRue 1989). For isotropic flow, we expect S(u) to be 0, and S(∂u/∂x) to approach a constant of -0.5 (Mohamed and LaRue 1989). Mydlarski and Warhaft’s studies indicate, by an uneven mean velocity profile, that homogeneity might be an issue for active grid flow (Mydlarski and Warhaft 1995). We expect the cross-wire steel meshes will assist achievement of homogeneous flow downstream, and create the appropriate conditions to study turbulence decay.