237order of magnitude as that of model A. Since buildings in model A have set-backs from the street, the open space between the second and third rows is wider than the corresponding space in model B. It would be presumed that ventilation in lightwells connected to the street space would have higher values in model A than in model B. Velocity measurements prove otherwise, however, as demonstrated in Fig. 3. It is further noticed in Fig. 3 that the velocity in the third row lightwell of model A decays with increasing free stream velocity until it vanishes. This interesting behaviour is attributed to vortices taking place in street spaces (Ghazi 1975).Model C was considered in order to study the effect of side passages. The only difference between models B and C is that the latter has no side passages. Measurements indicate that ventilation in the back space of the first side passage of model B is 12 times that of C. This is explained by the observation that the first side passage faces the wind directly, while the other passages are exposed to vortex flow, whereby they lose their effectiveness. Velocities at other locations in model B, however, are slightly lower than those in C. The indications are that, introduction of side passages significantly improves ventilation conditions in the first back space, whereas it adversely affects ventilation in the remaining rows.
ConclusionsComparison between airflow measurements on models A and B, which represent detached and attached buildings, respectively, shows that ventilation in back spaces between rows of buildings in model B is superior to that in the narrow back spaces of model A. On the other hand, high airflow is attained in the side spaces of model A. The side passages of model B considerably improve ventilation conditions in the first back space, while marginally affecting the conditions of rear rows of a group of buildings. Furthermore, the airflow in lightwells of model B is not significantly influenced by the presence of side passages.In conclusion, experimental results suggest that a combination of one frontal attached row with side passages, followed by rows of attached buildings with no side passages would be an arrangement superior to the layouts considered.
AcknowledgementsThis work was partly sponsored by Grant AT-5-79 of King Abdulaziz City for Science and Technology (KACST).
ReferencesAynsley, R. M.; Melbourne, W.; Vickery, B. J. 1977: Architectural aerodynamics. London: Applied Science Ghazi, M. A. 1975: Experimental investigation of the effect of surface discontinuity on step-backward flow. M.Sc. Thesis, Faculty of Engineering, Cairo University Givoni, B. 1976: Man, climate and architecture. 2nd edn. London: Applied Science Mohsen, M. A.; Olwi, I. A.; Ghazi, M. A. 1987: Aerodynamics and ventilation in buildings: Experimental investigation. Solar Wind Technol. 4, 501-507 Olgyay, V. 1963: Design with climate-bioclimatic approach to architectural regionalism. New Jersey: Prinston Univ. Press Olwi, I. A.; Ghazi, M. A.; Mohsen, M. A. 1988: Wind tunnel sim...