Numerical modeling of highly swirling flows in a through-flow cylindrical hydrocyclone

• Published in: AIChE journal Vol. 52; no. 10; pp. 3334 - 3344
• Main Authors: Ko, Jordan, Zahrai, Said, Macchion, Olivier, Vomhoff, Hannes
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc., A Wiley Company 01.10.2006; Wiley Subscription Services; American Institute of Chemical Engineers
• Subjects: Applied sciences; Cellulosa- och pappersteknik; Cellulose and paper engineering; Centrifugation, cyclones; Chemical engineering; Chemical process and manufacturing engineering; Engineering mechanics; Exact sciences and technology; Fluid dynamics; hydrocyclone separator; Kemisk process- och produktionsteknik; Kemiteknik; Liquid-liquid and fluid-solid mechanical separations; Numerical analysis; Reynolds stress transport models; Studies; swirling flow; TECHNOLOGY; TEKNIKVETENSKAP; Teknisk mekanik; Turbulence; Water flow; hydrocyclone separator; Turbulence; Turbulent flow; Swirling flow; Software package; Velocity distribution; Modeling; Reynolds stress transport models; Reynolds stress; Transport process; Hydrocyclone; Numerical simulation; Vortex; Axial flow; Separator
• ISSN: 0001-1541; 1547-5905; 1547-5905
• DOI: 10.1002/aic.10955

This article aims to identify the most appropriate numerical methodology for simulating hydrocyclone flows with high swirl numbers. The numerical results are validated against the tangential velocity measurements from a cylindrical hydrocyclone with a swirl number of 8.1, which is twice the typical swirl magnitude of industrial hydrocyclones. The linear and quadratic formulations of the Reynolds stress transport (RST) model are used to simulate the anisotropic swirling turbulent flow three‐dimensionally in the commercial software package Fluent™. The tangential velocity profiles predicted by the quadratic RST model are in good agreement with experimental data. They also show Rankine vortex patterns over the entire flow domain. In contrast, the linear RST model fails to predict this important swirl flow feature. In addition, both models predicted a complex axial flow reversal pattern not previously reported in hydrocyclones. This study clearly shows that the quadratic RST model is preferable for future hydrocyclone simulations, especially when the swirl number is large. All necessary physical and numerical parameters used to obtain converged results are given in this article. © American Institute of Chemical Engineers AIChE J, 2006