Search

Hygrothermal analysis of building components requires the knowledge of the convective heat and mass surface coefficients at exterior and interior building surfaces. Exterior surface coefficients are to a large extent influenced by the local wind speed near the surface. In most hygrothermal models, these coefficients are obtained from empirical formulae. They are used to relate the reference wind speed at a meteorological station to a local wind speed near the building surface and to relate the local wind speed to the heat surface coefficient. The vapor surface coefficients are generally determined from the heat surface coefficients using the Chilton-Colburn analogy. The existing empirical formulae for the heat surface coefficient as a function of wind speed however are based on only a limited number of measurements at a few facade positions and for a few building configurations. Little is known about the actual value and the variability of the surface coefficients across facades of different building geometries. In the past, most research on surface coefficients was experimental. Recently, numerical simulations based on Computational Fluid Dynamics (CFD) have been employed. Up to now, most simulations however have focused on interior surface coefficients. This paper presents high-resolution CFD simulations of exterior convective heat surface coefficients at the surface of a cubic building model. The simulations involve solving the entire structure of the boundary layer near the wall, including the logarithmic layer, the buffer layer and the laminar sublayer. Because of the high Reynolds numbers associated with wind flow around buildings and because the thickness of the boundary layer decreases with increasing Reynolds number, very high resolution grids, with cells as small as a few micrometers, have to be used near the building surface. The results show very distinct gradients across the facade. They indicate that the largest values of the surface coefficients occur at the edges, which are also the positions where wind-driven rain intensities are largest. This is of interest for the wetting-drying behavior of facades. Finally, it is shown that the low-resolution CFD approach, that has been used in some previous studies, and in which not the entire boundary layer is resolved, can yield some important discrepancies.

Convective surface coefficients are commonly used in whole building simulation to model the indoor and outdoor heat and mass transport boundary conditions. There are some standards and guidelines available for convective heat transfer coefficients, which generally include correlations for simple geometries. The convective vapour transfer coefficients are not so well documented and are often calculated with the heat and mass transfer analogy, and then subsequently applied for cases in which the analogy is not valid. As a step towards the development of guidelines on the use of convective vapour transfer coefficients in hygrothermal models, an experimental setup and numerical modeling results are presented. An experimental setup was designed for the purpose of validating two numerical models: 1) a coupled CFD-diffusion model and 2) a vapour diffusion model. In the CFD-diffusion model, heat and mass transport in the air domain is solved using CFD, while, in the material domain, vapour transport is modelled. This model allows the prediction of the convective vapour transfer coefficient. It was found that the vapour transfer coefficient changes in time evolving to constant values depending on air speed, and to a lesser degree on the relative humidity of the air. In the vapour diffusion model, a convective vapour transfer coefficient was imposed and the change in moisture content was determined. The experiment consisted of a wind tunnel placed in an environmental chamber with climate control capability. The air flow in the wind tunnel was driven by a variable-control fan that allows for a range of speeds in the laminar regime. Convective vapour transfer coefficients were determined using the experimental setup for a number of humidity and air flow conditions over gypsum samples. The results were compared with the numerical results for the purpose of validation and sensitivity analysis.

Copia digital. Madrid : Ministerio de Cultura. Subdirección General de Coordinación Bibliotecaria, 2007 La segunda secuencia de paginación (116 p.) corresponde al Appendix Las h. de grab. son litogr.: "Draw by Adam Neale, heath, sculp."