The suction induced by conical vortices on low-rise buildings with flat roofs

Abstract

Wind tunnel and full scale pressure studies of the flow over low-rise buildings have repeatedly shown that the largest mean, peak, and rms suction values observed anywhere on the building occur beneath the dual conical vortices that form along the leading edges of the roof during cornering winds. The manner in which these vortices cause extreme suction events is not well understood. In addition, pressures beneath the vortices display a range of unusual characteristics, such as the failure to adhere to quasi-steady theory, and discrepancies between wind tunnel and full-scale measurements. The objective of this study is to examine the process or "flow mechanism" by which a vortex instantaneously controls rooftop suctions. By better understanding the workings of the vortex itself, the upstream flow conditions which influence vortex behaviour can be directly connected to these suctions. This then permits the unusual pressure characteristics to be examined methodically. A unique experimental facility was developed to simultaneously record images of a vortex while measuring pressures beneath the visualization plane and flow velocities nearby. A new analytical model for the vortex pressure field was also developed and assessed experimentally. The model quantifies how two parameters, streamline curvature and flow speed above the vortex, control surface pressure. Experimental measurements confirm that the model accounts for changes in surface pressure with wind direction and proximity to the roof comer solely on the basis of these parameters. A full evaluation of the application of quasi-steady theory to pressures beneath the vortices is performed. Data gathered during the experiments and an analysis of the flow model show explicitly how the quasi-steady theory fails to accurately predict the condition and location of the vortex, though it does capture the essence of pressure changes due to vortex motion. Finally, the model suggests that by inhibiting the flow reattachment, the effect of the vortices on the roof can be eliminated. Several retrofit mitigation techniques that perform this role are assessed experimentally, and compared to the potential for suction reduction circumscribed by the model.