Modeling the Effect of Air Leakage in Hygrothermal Envelope SImulation

Vapor-tight wooden structures such as unvented flat roofs have acquired a bad reputation in Europe due to frequently reported moisture problems caused by vapor convection. While better detailing and workmanship may considerably improve the air-tightness of a roof assembly, field observations indicate that it is impossible to achieve a perfect air barrier under practice conditions. Therefore, a new European standard draft on wood protection specifies a convective moisture source for vapor control design analysis of building assemblies. This convective source is added as a safety margin to the amount of condensate caused by vapor diffusion when dew-point calculations are performed.

The paper describes how this concept of convective moisture source is translated into an air-leakage model for hygrothermal simulation tools. Since the bulk of exfiltrating air is flowing straight through larger gaps and joints, it is unlikely to do any harm because the flow channels will generally become to warm for vapor condensation. Therefore, the model assumes that only small leaks with tortuous paths contribute to the convective moisture source. The challenge is to determine the flow rate through the small moisture-relevant leaks. Based on field tests and theoretical assumptions a small leak air permeance is defined that serves to calculate the convective moisture entry. The resulting flow rate depends on the air pressure differentials due to stack effect and mechanical ventilation. Wind induced pressure differentials are neglected because they are very transient in nature (changing force and direction) and more complex to determine. After specifying the most likely position for convective condensation within the building assembly, the moisture source is calculated hourly depending on the indoor and outdoor climate conditions.

The convective moisture source model has been validated by comparison with field tests. Applying the model offers the possibility to assess the risk of moisture damage caused by vapour convection. It demonstrates that flat roof assemblies with vapour barriers are more prone to moisture problems than those with moderate vapour retarders, which is in line with practical experience. The model also indicates the limits of moisture removal by vapour diffusion of building assemblies subject to vapor convection.