Wall shear flows

© 1985 Dr. Kong Loong LimA short literature survey outlining recent developments in the modelling of wall turbulence is presented. One of the most recent developments, the A-vortex model of Perry and Chong (1982), which was based on Townsend's (1976) attached eddy hypothesis, is discussed. The use of hot-wires, in particular X-wires, for measuring turbulence quantities, has been found to have several limitations especially over rough surfaces. Some methods of overcoming these shortcomings are proposed, tested and discussed. Mean flow, broad-band turbulence intensity and spectral measurements were carried out over smooth and rough surfaces in zero pressure gradient boundary layers. The spectra measured in the turbulent wall region were found to follow the scaling laws put forward by Perry and Abell (1977) and the A-vortex model of Perry and Chong (1982) for both smooth and rough walls. The A-vortex model of Perry and Chong was used to generate spectral distributions corresponding to some flow cases tested and agreed well with the measured spectral distributions. Strong evidence is presented for the existence of an inertial subrange, where the small scale motions are locally isotropic. The spectral laws put forward by Perry and Abell (1977) are used to infer the broad-band turbulence intensity distributions and reasonable agreement was shown with the measured distributions. The measured broad-band turbulence intensity distributions strongly support the Townsend (1956) Reynolds number similarity hypothesis and the extended form of this hypothesis for rough surfaces of Perry and Abell (1977), at least for the limited range of Reynolds numbers tested here. The results also strongly support the theory that the structure of wall turbulence consists of eddies "attached" to the wall, and that their distribution in the boundary layer follows that proposed in the model of Perry and Chong (1982)