Name: MICHEL DE OLIVEIRA DOS SANTOS
Publication date: 20/10/2025
Examining board:
Name |
Role |
|---|---|
| RUDOLF HUEBNER | Examinador Externo |
Pages
Summary: Attenuators are widely employed in industrial pumping systems to reduce pressure and flow pulsations that cause excessive noise, mechanical vibrations, and process instabilities. Conventional devices use compressed gases (typically nitrogen or dry air) as the deformable medium; however, aiming at greater reliability, an attenuator may operate without gas compression provided that satisfactory attenuation is maintained within the desired frequency range. While three-dimensional numerical analyses capture behaviors beyond simplified models, lumped approaches remain useful for identifying global trends with low
computational cost. In this work, a lumped-parameter model was developed to estimate the attenuation of intermittent-flow amplitude in a gas-free tubular attenuator. The device was also analyzed using a three-dimensional fluid–structure interaction (FSI) model under laminar flow. Experimentally, two tubular attenuators with different compliances were tested in laminar and turbulent regimes, at several pulsation frequencies and in two configurations: inline and side-branch. The analytical solution of the two-element Windkessel model showed that attenuation depends on the dimensionless frequency () and on the system parameter KRCL, remaining independent of the mean-flow Reynolds number (Reta). A modified Windkessel formulation was proposed to incorporate the dependence on Reta under turbulent conditions. FSI simulations performed for two values
of KRCL (1.75 × 104 and 3.50 × 104), two values of Reta (750 and 1500), and six values of (20–160) confirmed qualitative agreement with the analytical model while revealing additional sensitivity to Reta associated with the mean internal pressure. In the turbulent experiments (17140 < Reta < 45700, 168 < < 450), attenuation increased with both parameters. The attenuator with Dd = 36 mm reached up to 85%, whereas the one with Dd = 32 mm reached 75.6% but showed negative attenuation (down to –36%) under some operating conditions, attributed to its lower compliance and to geometric and constitutive nonlinearities. The relation between attenuation levels did not scale linearly with the ratio of compliances, underscoring the complexity of the phenomenon. In the laminar experiments (865 < Reta < 2212, 8.9 < < 24.5), attenuation ranged from 17.9% to 78.1%, with no negative cases, and similar results were obtained for both configurations.
Across regimes, the trend of increasing attenuation with and Reta persisted, indicating that performance is governed primarily by the compliance response to pressure and flow fluctuations rather than by the flow regime itself.
