Influence of Turbulent Flow Characteristics and Coherent Vortices on the Pressure Recovery of Annular Diffusers: Part B — Scale-Resolving Simulations

Abstract

This paper examines the diffuser flow with consideration to turbine outflow conditions. The setup consists of a low-speed axial diffuser test rig, that represents a 1/10 scaled heavy-duty exhaust diffuser with an annular and a conical diffuser part. In part A of this paper it was shown through experimental investigation that the turbulent kinetic energy as well as the Reynolds shear stresses are the relevant physical parameters that correlate with diffuser pressure recovery. To complement the experimental investigations, unsteady scale-resolving CFD simulations are performed, applying the SST-SAS turbulence model. As a first step, the numerical approach is validated by means of the experimental data with regards to the diffuser’s integral parameters as well as the prediction of local flow characteristics. In a second step, the interaction of coherent vortices generated by the rotor and the diffuser’s boundary layer are analyzed by means of the validated SST-SAS results. These vortices are found to have a major impact on the boundary layer separation in the region immediately downstream of the rotor and at the diffuser inlet.

Correlation between total pressure losses of highly loaded annular diffusers and integral stage design parameters

“…Kuschel and Seume (2011) conducted additional experiments using a NACA profile instead of cylindrical spokes. A detailed analysis of these experimental results can be found in Kuschel et al (2015) and Drechsel et al (2015). They conclude that the increase in pressure recovery is caused by the interaction of tip vortices of the rotor blades with the diffuser boundary layer.…”

On the design of small to medium scale radial inflow turbines for supercritical CO2 power cycles

“…Contrasting the inlet delivery system, interactions between components are significant in determining performance and optimum geometry of the diffuser. Recent studies [78,79,80,81,82] indicate that rotor -diffuser interactions are significant, and that fully coupled, URANS stage simulations are necessary to resolve diffuser component performance. The use of such a modelling framework leads to more compact, higher performing diffuser geometries than would otherwise be designed had the interactions not been accounted for.…”

“…RANS calculations are used to determine appropriate mesh density and to initialise URANS calculations, whilst URANS calculations are used for performance calculations. RANS equations are closed with the SST k-ω turbulence model [74], whilst the anisotropic SAS SST turbulence model is used for URANS cases [75,76], as used in prior numerical diffuser studies [80,81,82].…”

“…As reviewed in Section 7.2, this appears to be the preferred level of simulation com-plexity for assessing diffuser performance under influence of rotor exit conditions, however, results can be obtained much more rapidly from steady state (RANS) calculations. By contrast, if statorrotor interactions persist into the rotor as suggested by some recent experimental campaigns [82], it may be necessary to model a non pitch matched stator and rotor, i.e. Zs = 21, Zr = 16.…”

“…Additional momentum close to the shroud surface, afforded by the tip gap, supports the boundary layer against the adverse pressure gradient within the diffuser and enables efficient operation of a more compact diffuser. Further studies [80,81,82] have also highlighted the need for anisotropic turbulence modelling, and full stage URANS calculations in order to adequately model diffuser inlet flow distortions and hence reasonably predict diffuser performance.…”