The paper presents an original computer code for the analysis of contaminant diffusion in rooms developed at the Politecnico di Torino and its experimental validation by means of a test facility located at the University of Basilicata (Potenza). Thevelocity fields in isothermal conditions, together with localages of the air, have been analysed and compared, with different ventilation strategies and number of air changes.

Application of hot wirelfilm ariemometry in room air flows presents difficulties because: (1) the effect of natural convection due to the heated wire beconies significant for low air velocity measurements; (2) the angle sensitivity of a hot wire becomes small at low air velocities, which makes it difficult to resolve the direction of each velocity component. This study aimed at quantifying the uncertainty of tlie hot wire anernornetry and examining the angle sensitivity of a hot wire in low air velocity measurements.

A new simple method is proposed here for the experimental singling out of the air flow patterns in a room. It is based on the use of a series of thermofoil probes, arranged in a lattice, that can be suspended at the ceiling of the room under test.

A set of diagrams for estimating flow coefficients and exponents in the power law flow equation for cracks are presented. The diagrams are primarily intended for those who perform infiltration calculations by hand or by using a computer program for single and multi-zone infiltration and ventilation calculations. The error introduced by the estimation technique is compensated for by means of a correction coefficient with aspecific value in different pressure difference intervals. A computer program performing the calculations behind the diagrams is available for public use.

The basis of this study are experimental results obtained on a real scale cell in controlled climatic conditions which are used to show the potential influence of radiative participation of inside air on natural convection in a room. In a second part, a numerical analysis of flow patterns and heat transfer in a twodimensional thermally driven cavity containing a participating fluid is presented.

Both infiltration and exfiltration has a predominant influence on the space heating requirements in cold climates. Good predictive design methods are required to estimate the air leakage component in buildings. This predictive methods will be useful in implementing the air leakage control strategies for reducing the problems associated with air infiltration in both new and existing high-rise buildings.

This paper is concerned with the application of air flow simulation in design. It describes the real world application - and the results of this with respect to building design improvement - of a building energy modelling system, ESP (RT) , which supports the analysis of coupled heat and fluid flow as encountered in a building andlor plant environment. The use of the system, and the design benefits to accrue, are demonstrated by elaborating two real world case studies.

Due to the limitations of computer storage and time the flow boundary conditions at an air inlet device have to be specified for numerical simulations of air flow patterns in rooms. With regard to this the present work gives velocity measurements near an industrial air inlet using a Laser-Doppler-Anemometer. From the stochastic velocity data the time-averaged velocity components, standard deviation and turbulent kinetic energy are evaluated.

The paper discusses methods to set boundary conditions at the air supply opening in predictions of room air flows with computational fluid dynamics. The work is a part of the International Energy Agency project "Air Flow Patterns within Buildings", Annex 20. The air supply terminal in the Annex 20 project is a commercial diffuser which creates a stagnation region and a complicated wall jet below the ceiling. Fairly well predictions in the wall jet region were obtained replacing the diffuser by a simple opening which has the same momentum flow as in the diffuser.

Recent full scale experiments has detected the presence of low Reynolds number effects in the flow in a ventilated room. This means that one are unable to predict the flow patterns in some geometries for air change rates - or Reynolds numbers - which are relevant for ventilation engineering by a standard model of turbulence. In this paper it is investigated if it is possible to simulate and capture some of the low Reynolds number effects numerically using time averaged momentum equations and low Reynolds number k-e model.