Characterisation and simulation of a general ventilation filter based on a double porosity approach

The air filters installed in ventilation systems are designed for collecting dust and particles. These filters usually consist in folded paper or textile, arranged either as a flat slab or in a V-shape frame. Some of them may also contain activated carbon in order to remove molecular compounds. Experimental results show that they can significantly attenuate the upstream sound sources such as fans or other regenerated noise. Though, these acoustic properties are not strictly speaking designed and often result from the tailored filtering capacities. The present work proposes a three-stage methodology to design and optimize the acoustic performance of these filters. The first stage is the characterization of the intrinsic acoustic parameters of the paper or textile. The second stage consists in the validation of these parameters regarding sound attenuation as measured in impedance tube. This includes analytical and finite element computations to correctly account for the mounting conditions of the specimen in the impedance tube. The final stage consists in predicting the sound attenuation measured on the full-size filter, set in shape using the porous composite material theory. The obtained results provide a satisfactory correspondence allowing for an accurate design at an early stage of the project.