The Effects of Geometry on Flexible Duct CFD Simulations

Abstract

Flexible ducts have been widely used in the building industry due to low cost and ease of installation. These ducts can be installed in a wide range of configurations, which creates a challenge for pressure loss calculations. Computational fluid dynamics (CFD) simulations allow variable configurations and are emerging as an alternative to laboratory measurements. Issues with the CFD simulations of flexible ducts have been modeling the complex geometry and the computational requirements to complete a simulation. In this study, five 8” diameter 15% compressed duct geometries were modeled including: periodic-triangular (PT), helix-triangular (HT), periodic-circular (PC), helixcircular (HC), and periodic-double-triangular (P2T). These modeled duct shapes were compared to determine the complexity of modeling and computational requirements. The performance of each model was determined based on the agreement with the measured data. The difference of static pressure differentials between PT and HT geometries was within 3%. Similarly, static pressure differentials between PC and HC, geometries were also within 3%. These results suggested that simulating the helix, which is the actual geometry of flexible ducts, had negligible effect on the results. In addition, simulation results of the models with triangular wall geometry were within 50% closer agreement to measured data than the circular wall geometry. The results suggested that periodicdouble- triangular (P2T) geometries, which are also more computationally efficient, can be used in 8” diameter 15% compressed flexible duct simulations. The calibrated CFD model can then be used for various duct configurations.