Investigation of a vortex tube using three different RANS-based turbulence models

Abstract

The main objective of this study is to capture the function of temperature separation inside a vortex tube (VT), firstly, via testing it at different inlet pressures (4, 5, 6, and 7 bar) and then examining the best performance of the VT through computational fluid dynamics (CFD) along with three various Reynolds-averaged Navier–Stokes-based turbulence models, i.e., standard $$k - \varepsilon$$ , standard $$k - \omega$$ , and shear-stress transport $$k - \omega$$ . Moreover, a comparison between CFD outputs and experimental data, at the inlet pressure of 7 bar, demonstrates that the standard $$k - \varepsilon$$ model outperforms other turbulence models utilized in this study. The internal flow behavior is also examined for further illustration of its features via CFD. It is reported that a convergent–divergent virtual duct is formed through the streamline at the region wherein injected flow (from the inlet nozzles into the vortex chamber) swirls toward the cold orifice, resulting in an expansion. Consequently, the Mach number goes above one and temperature drops. Thus, the sudden expansions alongside free and forced vortices as well as the secondary circulation flow have a significant impact on energy separation in the VT.