Crack-width-dependent modelling of air infiltration under low pressure differences: Experimental and computational fluid dynamics-based insights

Air infiltration through cracks in building envelopes significantly affects indoor conditions and energy performance under natural pressure differences. However, the applicability of commonly used infiltration models across varying crack widths remains unclear. This study investigates air infiltration through single straight cracks with different widths under pressure differences ranging from 0 to 10 Pa using static pressure chamber experiments and computational fluid dynamics (CFD) simulations. Experimental results were used to evaluate the performance of the power-law and quadratic models, while CFD simulations provided insight into pressure distribution and flow characteristics. The results show that model performance depends strongly on crack width. For crack widths ≤1.5 mm, the quadratic model provides reliable predictions. As crack width increases beyond this range, the power-law and modified quadratic models show improved performance, while the conventional quadratic model exhibits increasing deviation. Furthermore, for larger crack widths, the friction-related term in the quadratic model becomes negligible. Based on this observation, a modified quadratic model is proposed, which avoids nonphysical predictions while maintaining high accuracy; for crack widths ≥2.0 mm, it achieves coefficients of determination above 0.99 and prediction errors below 2%. This study establishes a crack-width-dependent framework for infiltration model selection under low pressure differences.