Moisture response of building facades to wind-driven rain: Field measurements compared with numerical simulations

• Published in: Journal of wind engineering and industrial aerodynamics 2009-08, Vol.97 (5), p.197-207
• Main Authors: Abuku, M., Blocken, B., Roels, S.
• Format: Article
• Language: English
• Subjects: Applied sciences; Buildings; Buildings. Public works; Climatology and bioclimatics for buildings; Computation methods. Tables. Charts; Computer simulation; Convective moisture transfer coefficient; Driving rain; Exact sciences and technology; External envelopes; Facades; Flux; Ham; HAM analysis; Mathematical models; Moisture; Moisture transfer; Numerical simulation; On-site experiment; Rain; Raindrops; Rainwater uptake; Structural analysis. Stresses; Validation; Wall. Partition; Water drop impact
• ISSN: 0167-6105; 1872-8197
• DOI: 10.1016/j.jweia.2009.06.006

Wind-driven rain (WDR) is one of the most important boundary conditions governing the hygrothermal behaviour of building facades, which is usually numerically analysed with the so-called Heat-Air-Moisture (HAM) transfer models. In the traditional approach of HAM transfer models, WDR is implemented in a simplified manner: the total mass of all raindrops impinging on a certain surface area of a building facade during the time interval of the meteorological input data (typically 1 h) is spatially and temporally averaged and is supplied to the facade as an averaged moisture flux. However, real WDR is the sum of individual raindrops that impinge on the facade in a spatially and temporally discrete modus, and that do not only spread at impact, but may also splash or bounce off the facade. Therefore the reliability of this simplification can be questioned. To investigate its validity, a new experimental set-up was developed at a full-scale test building. It allows simultaneous and continuous measurements of the reference wind speed and direction, WDR intensity, outdoor air temperature and humidity, as well as the response of facade material samples to these environmental conditions. For this purpose, a measuring device was developed that monitors the weight change of the sample with a resolution of 5 mg. Temperatures at the interior and exterior material surfaces are also monitored. The whole measurement data set is used to check the validity of the traditional numerical approach. Large differences are found between the measurement and simulation results, which cannot solely be attributed to the uncertainty in the convective moisture transfer coefficient, but may be due to two additional reasons: the occurrence of splashing and bouncing at raindrop impact on the facade, which is not included in the model, and/or errors in surface moisture evaporation and absorption due to modelling the actual random and discrete raindrop impingement as a simplified averaged moisture flux.