ENGINEERING AND CFD

Pumps are used in many technical, industrial and household applications for conveying any kind of fluids, including multi-phase flow. Due to increasing requirements on a flexible operation, pumps are often run at off-design operation, where for example by legislature, demands on efficiency are continuously increasing. The optimization of high-efficiency pumps can only be achieved by modern computational fluid dynamics methods (CFD) in combination with accurate turbulence models. While for the design point operation of centrifugal pumps, where an essentially steady flow field is present and statistical turbulence models yield an appropriate prediction of the characteristics, the flow field gets increasingly unsteady towards off-design operation.

Special designs as e.g. sewage pumps are characterized by a single-blade impeller and show significantly unsteady characteristics even in the design point [1]. The same holds for positive displacement pumps where due to the periodic piston oscillation, the flow field is inherently unsteady and turbulent.

For such highly-unsteady and turbulent flow fields, statistical turbulence models tend to fail. On the other hand, Large-Eddy Simulation (LES) models, in which the large-vortex part of the turbulent spectrum is directly resolved, show a much better flow prediction. However, LES requires the direct resolution of the major part of the turbulent kinetic energy and, for the high Reynolds numbers encountered in pump flows, the spatial resolution and thus computational effort are tremendously high, even on modern HPC systems. Therefore, an assessment of scale-adaptive turbulence simulation (SAS) models is provided, which recover a statistical flow solution in regions of low unsteadiness and – like Large-Eddy Simulation – resolve a part of the turbulent spectrum down to the available grid resolution for highly unsteady flow regions. An in-depth analysis of the SAS model and a comparison with a LES reference solution on simplified benchmark test cases on separated flow [2] reveals that the SAS provides a significantly improved flow solution compared to an under-resolved LES.

An application of the SAS on a centrifugal pump flow and comparison to measurement data [2,3,4,5] confirms its beneficial properties for high Reynolds number pump flow. Time and space-resolved measurement data is available in the highly unsteady impeller–stator flow interaction region for validation. In Figure 1, the research pump configuration is demonstrated by its meridional impeller view (Figure 1 a), an axial view of the volute casing with the circumferential coordinate ε  (Figure 1 b) and the time coordinate x/t , which corresponds to the time range when one impeller blade spacing passes the measurement position (Figure 1 c).

The ensemble-averaged mean flow angle α  and turbulence intensity TI  is illustrated in Figure 2 in the measurement plane for part-load operation. While α  is well predicted even with a conventional statistical turbulence model (STM), the TI  is significantly overpredicted, in particular for the circumferential position ε= 0°  close to the tongue. By the SAS on the other hand, even the TI  is well predicted for any circumferential location. This example flow configuration demonstrates that by the SAS, turbulence-scale resolving capabilities are retained even by the complex high Reynolds number flow, and the turbulence fluctuations in terms of the TI  are accurately predicted, which has also been demonstrated by different rotating machinery applications as e.g. pump mixers [6].