Sensitivity analysis of orifice length of micro-cavity array for the purpose of turbulence attenuation

The micro-cavity array is a unique surface geometry which aims to passively dampen the total energy of coherent motions within the near-wall region of a turbulent boundary layer. Previous works have shown the effectiveness of this device, in particular the reduction of sweep and turbulence intensities by 13% and 14%, respectively. Preliminary parametric studies and investigations have been made on the cavity array, particularly regarding its geometry and configuration of the backing cavity, however there are still gaps in the knowledge of ideal geometric features of the cavities themselves. This paper presents a parametric study to investigate the sensitivity of the micro-cavity array’s ability to attenuate turbulence within the boundary layer to the aspect ratio of the orifices installed (length on diameter of orifices) and to further understand the physics behind the flow effected by the array itself. Five distinct orifice aspect ratios were investigated using constant temperature anemometry in a zero-pressure-gradient boundary layer flow. The velocity profiles generated were compared to a canonical boundary layer flow, and a relationship between the reduction in turbulence statistics and aspect ratio was constructed. Reductions in turbulence intensity were greater as the aspect ratio of the orifices increased, with a maximum reduction of 12.9% for a momentum-based Reynolds number of Re θ = 3283 and orifice length of l + = 534 (normalized by viscous wall units). When observing a fixed wall height, both sweep intensity and streamwise turbulent kinetic energy supported the turbulence intensity results. The pre-multiplied energy spectra were reduced by a maximum of 11% for Re θ = 4651 at the largest orifice length tested ( l + = 763 ). The results from this paper indicate that the orifices of a micro-cavity array should be as large as possible, however this claim is only true for the orifice lengths investigated in this paper ( 105 < l + < 763 ). Graphical abstract