Swirling diffuser flow

Underlying Flow Regime 4-06 © copyright ERCOFTAC 2004

Best Practice Advice

Best Practice Advice for the UFR

Based on the study test case and a review of related studies it is possible to make the following recommendations for best practice modeling of swirling diffuser flow.

Key Physics

Turbulence and the effect of swirl on turbulence

Boundary layer flow and separation

The flow structure also depends on degree of swirl, angle of divergence and the ratio of the upstream to downstream flow area.

Generality of Advice Given

The best practice advice is based on data collected from numerous computations of the study test case.

Eleven of these simulations were submitted as part of an ERCOFTAC workshop entitled “Data Bases and Testing of Calculation Methods for Turbulent Flows” in Karlsruhe from April 3 to 7, 1995.

The advice applies to diffusers operating with low to moderate amounts of swirl; it was derived from tests with an inlet swirl number of order 0.35, and on this basis should not be assumed to apply at much higher swirl numbers. In particular, if flow reversal occurs in the solution along the centre-line of the diffuser, the inlet swirl can no longer be regarded as moderate.

Numerical Issues

Discretisation — use a higher order scheme (second order or above) for momentum equations.

Grid and grid resolution — near wall approximations have major effects on the flow field [9] and hence the computational mesh must be of sufficient resolution near the wall to support the near wall approximation implemented. If wall functions are used, ensure that the near wall mesh is not refined beyond the limit of their validity based on the y+ values. Where possible, quad or hex elements should be used.

Boundary Conditions and Computational Domain

Results are sensitive to the distributions of inflow variables: therefore, use the correct (experimentally derived) distributions of inlet axial and swirl velocity if available. If the inlet distributions are not known, use distributions that are characteristic of the origin of the swirl; for example, a solid body (forced vortex) swirl velocity distribution if the swirl is generated by a long rotating component.

The radial pressure distribution in a swirling flow is non-uniform: do not impose a uniform pressure outlet boundary condition.

Physical Modelling

Turbulence — Overall the standard k-ε model with wall functions gave results which, apart from an insufficient reduction in the center line velocity, are reasonable for all quantities that are of practical interest. For the study test case no significant advantage was shown by the ASM or RSM models. However, the swirl may not have been strong enough in this case to show the superiority of Reynolds stress models reported in the literature for swirling flows.

Near Wall Modelling — If wall functions are used the near wall mesh should only be refined to the limit of their validity based on the y⁺ values.

Recommendations for Further Work

In the study test case, flow reversal at the diffuser outlet did not occur. Further work should be carried out to identify appropriate boundary conditions and domain extents at the diffuser outlet for systems where flow reversal at the outlet does occur.

Also, further studies should be carried out at higher inlet swirl numbers to determine whether ASM or RSM type turbulence models yield more accurate predictions than the more standard isotropic eddy viscosity k-ε type turbulence models.

Contributors: Chris Carey - Fluent Europe Ltd