Similitude Problems in Building Aerodynamics

Use of model techniques for determining pressure distribution and wind loads on buildings is considered. None of existing simplified laws adequately accounts for velocity distribution and roughness effects. A more complete relation is derived and verified in a series of tests. A cube placed in a boundary layer of varying thickness is used to demonstrate the variation of pressure distribution produced by these effects. Uniform velocity test data reported in the literature is examined and found to contain large discrepancies. Wind tunnel verification tests show these to be caused by a number of deficiencies in model mounting arrangements. The model must be mounted on a ground plate with a long downstream section and gaps which allow air flow into the wake must not exist. A Reynolds number influence on the pressure distribution is also demonstrated.