Discusses the two methods for calculating air infiltration given in the ASHRAE handbook. These are the air change method, a gross estimate based on the number of windows and doors in each room, and the crack method based on measurements of flow through the cracks around windows and doors. Presents comparisons of tracer gas measurements with calculations by both air change and crack methods for test houses in California and Minnesota. Find agreement is adequate for sizing equipment but that the crack method underestimates infiltration at low wind velocity.
Describes apparatus and experimental techniques for full and model scale measurements on test buildings. Discusses "blocking effect" of a large model in a small tunnel. Shows that model law derived in part one is valid forphenomena dependent on wind velocity. Compares model to full-scale tests. Discusses air flow around a house, pressure on walls and different types of roofs
Discusses the physical nature of atmospheric boundary layer flows. Concludes that the primary aims in the simulation of these flows in a wind tunnel should b to model the relevant scales and intensities of turbulence. Simulation of the variation of mean wind velocity with height is also desirable. Proposes a system of barriers and vortex generators as a means of simulating turbulent and neutral atmospheric boundary layers.Discusses the characteristics of induced turbulent flow. Finds that the developed flow is a good approximation to the natural boundary layer.
Reports an investigation of wind loading with emphasis on the local pressure fluctuations, on a small scale building model in a thick turbulent boundary layer wind tunnel. A striking similarity between the oncoming turbulent energy spectra andsurface pressure-fluctuation spectra was consistently observed. This similar behaviour suggests that the upstream turbulence plays a dominant role in producing the pressure fluctuations on the upwind face of a bluff body.
Discusses the effect of wind on air change rates in buildings. Reports series of model tests conducted in a water flume and a wind tunnel. A plexiglass box with holes in it was filled with gas, either nitrogen or carbon dioxide, and placed in a controlled air flow. The concentration of gas was plotted in a semi-logarithmic form. Gives typical examples of these graphs.Discusses feasibilty of estimating rate of air change by a hyperbolic function, but finds that more tests are needed forpractical recommendations.
Discusses conditions that must be satisfied for a model in a wind-tunnel to give the same air-flow as a full-sized building. Reports two series of tests on interior and exterior air flow patterns, made on a full-sized building and a scale model of the building. Air flow patterns were observed using titanium tetrachloride smoke. Tests were also made to determine the limits by which the product of the height of the model by the air speed may vary without serious error.
Reports measurements of pressure distribution on square cylindrical models in wind tunnel. Vertical distribution of wind velocity was produced by grids of horizontal rods at varying spacing. Wind pressure distributions on model-scale buildings were obtained, varying the height, width, depth and winddirection. To compare results, a large-scale model 3.6 metreshigh 1.2 metres by 1.2 metres in plan was placed on the shore and pressure distribution measured during a strong wind. Gives diagrams of pressure distributions.
Describes tests made to find wind pressure on models in a low-velocity wind tunnel. Three basic forms:- a semi-cylinder, a rectangular vertical wall and a block-type gabled building were tested at several different angles to the wind. Gives typical pressure patterns for block-type model. Suggests use of average pressure coefficient for calculation of wind loads determined from pressure distribution. A short series of tests on the effect of shielding building showed that negative pressure on some walls could be increased by an adjacent building.
Detailed sets of time-averaged surface pressure coefficients were recorded over the walls and roof of a rectangular building model, set in a simulated high density urban area. The 1/400 scale model represented a generalized smooth surfaced building of 100x150 ft plan form, whose height varied from 300 ft to 200,100 and 50ft. Surrounding roughness elements equalled the heightof the 50ft, building model. Tests were carried out at twelve wind angles using a power low velocity profile with an exponent of 0.43.
Sets out the similarity requirements which must be observed so that results of wind tunnel tests may be used to predict behaviour of full-scale prototypes in the natural wind. Discusses rigid models, suspension bridges, models of slender towers. Outlines problems of representing natural wind and the effect of wind tunnel-wall interference. Introduces correction procedure for wake blockage which permits the use of larger wind-tunnel models than would otherwise be possible without serious errors.
Login