Reduction of turbulent skin-friction drag by passively rotating discs

TL;DR

This study numerically investigates how passively rotating discs attached to walls can reduce turbulent skin-friction drag, showing up to 5.6% reduction with certain configurations and providing insights into the fluid-disc interaction.

Contribution

It introduces a coupled fluid-disc model to analyze passive disc rotation effects on turbulence and drag reduction, with new findings on optimal disc sizes and partial surface effects.

Findings

Maximum drag reduction of 5.6% over the entire wall.

Large reduction in turbulence activity on the spinning portion.

Drag reduction depends on wall-slip velocity and shear stress distribution.

Abstract

A turbulent channel flow modified by the motion of discs that are free to rotate under the action of wall turbulence is studied numerically. The Navier-Stokes equations are coupled nonlinearly with the dynamical equation of the disc motion, which synthesizes the fluid-flow boundary conditions and is driven by the torque exerted by the wall-shear stress. We consider discs that are fully exposed to the fluid and discs for which only half of the surface interfaces the fluid. The disc motion is thwarted by the fluid torque in the housing cavity and by the torque of the ball bearing that supports the disc. For the full discs, no drag reduction occurs because of the small angular velocities. The most energetic disc response occurs for disc diameters that are comparable with the spanwise spacing of the low-speed streaks. A perturbation analysis for small disc-tip velocities reveals that the two-way nonlinear coupling has an intense attenuating effect on the disc response. The reduced-order results show excellent agreement with the nonlinear results for large diameters. The half discs rotate with a finite angular velocity, leading to large reduction of the turbulence activity and of the skin-friction drag over the spinning portion of the discs, while the maximum drag reduction over the entire walls is 5.6%. The dependence of the drag reduction on the wall-slip velocity and the spatial distribution of the wall-shear stress qualitatively match results based on the only available experimental data.