Effects of the yaw angle on the drag reduction rate of riblets were investigated experimentally for trapezoidal riblets with ridge angle of \(30^\circ\) and height-to-spacing ratio of 0.5 in a turbulent wind channel at Reynolds numbers (based on the friction velocity and channel half depth) \(\textrm{Re}_\tau = 520 - 2100\) . Drag reduction rates were estimated carefully by comparing the streamwise pressure gradients between turbulent channel flows with and without riblets. The maximum drag reduction rate which was about 7% for the streamwise riblets was maintained without any attenuation for the yaw angle \(\phi\) below \(10^\circ\) . When \(\phi\) exceeded \(10^\circ\) , the drag reduction rate was attenuated steeply and the performance of riblets was almost lost at \(\phi = 15^\circ\) . The yaw angle at which the drag turned to increase was by 10º smaller than the value reported for the saw-tooth riblets. The yaw angle effect on the riblet performance was further examined by detailed comparisons of mean velocity profiles and turbulent structures near the ribbed surfaces for \(\phi = 10^\circ\) and \(15^\circ\) . Although no appreciable difference in the turbulent intensity and pre-multiplied spectrum of near-wall turbulence was found between these cases ( \(\phi = 10^\circ\) and \(15^\circ\) ), the mean velocity profile for the case of \(\phi = 15^\circ\) indicated an evidence of flow separation at the ridges of riblets, leading to the attenuation of drag-reduction rate due to increase in the pressure drag.
