Experimental Investigation of Tornado-Induced Pressures on Low-Rise Buildings

Abstract

Tornadoes pose a significant danger to human life and structures. Research regarding the effects of tornado-induced loads on residential buildings is in incipient stages and there are no specialized construction standards in place to recommend criteria applicable to structures for withstanding tornadic winds. Three residential house models with different geometries were tested in the Wind-induced Damage Simulator (WDS) built at the University of Ottawa. The WDS is capable of simulating pressures induced by multidirectional and tornadic winds. The peak pressure coefficients were calculated on the walls and roofs of the houses and an analysis was performed on the effects of house model orientation, roof pitch angle, and exposure duration. The peak pressure coefficients were then compared to the NBCC 2015 code to clarify if there were any limitations of the current wind design criteria. It was found that the building orientation did not have a significant effect on pressure coefficient trends and magnitudes on the walls and roofs. For the low roof pitch angle models, it was noticed that the suction on the roof was much greater than the higher roof pitch angle models. An interesting observation was made that found that the leading edge of the walls in the direction of the clockwise tornadic flow were always under greater suction than the trailing edge, which causes a torsional effect on the entire model. When comparing the peak pressure coefficient values to the NBCC 2015 recommended values for the secondary cladding members, it was found that the CpCg stipulated in the code were similar to the experimental tornado Cp’s for the walls. However, the Cp’s on the roof were much greater in the experiments when compared to the NBCC 2015. The CpCg of Zones S and Zone R, which are the edges and central regions of the roof, greatly exceed the minimum values in the NBCC 2015. More experiments for residential house models of different geometries should be conducted in order to propose new tornado-induced pressure coefficients to be used in the design of the structure located in tornado-prone areas.Tornadoes pose a significant danger to human life and structures. Research regarding the effects of tornado-induced loads on residential buildings is in incipient stages and there are no specialized construction standards in place to recommend criteria applicable to structures for withstanding tornadic winds. Three residential house models with different geometries were tested in the Wind-induced Damage Simulator (WDS) built at the University of Ottawa. The WDS is capable of simulating pressures induced by multidirectional and tornadic winds. The peak pressure coefficients were calculated on the walls and roofs of the houses and an analysis was performed on the effects of house model orientation, roof pitch angle, and exposure duration. The peak pressure coefficients were then compared to the NBCC 2015 code to clarify if there were any limitations of the current wind design criteria. It was found that the building orientation did not have a significant effect on pressure coefficient trends and magnitudes on the walls and roofs. For the low roof pitch angle models, it was noticed that the suction on the roof was much greater than the higher roof pitch angle models. An interesting observation was made that found that the leading edge of the walls in the direction of the clockwise tornadic flow were always under greater suction than the trailing edge, which causes a torsional effect on the entire model. When comparing the peak pressure coefficient values to the NBCC 2015 recommended values for the secondary cladding members, it was found that the CpCg stipulated in the code were similar to the experimental tornado Cp’s for the walls. However, the Cp’s on the roof were much greater in the experiments when compared to the NBCC 2015. The CpCg of Zones S and Zone R, which are the edges and central regions of the roof, greatly exceed the minimum values in the NBCC 2015. More experiments for residential house models of different geometries should be conducted in order to propose new tornado-induced pressure coefficients to be used in the design of the structure located in tornado-prone areas.