0
Research Papers: Flows in Complex Systems

PIV Measurements and CFD Computations of Secondary Flow in a Centrifugal Pump Impeller

[+] Author and Article Information
R. W. Westra1

Department of Mechanical Engineering, University of Twente, 7500 AE Enschede, The Netherlandsremko.westra@cdsengineering.nl

L. Broersma2

Pump Division, Flowserve Corporation, 7553 AE Hengelo, The Netherlandsl.broersma@suzlon.com

K. van Andel

Department of Mechanical Engineering, University of Twente, 7500 AE Enschede, The Netherlandsk.vanandel@utwente.nl

N. P. Kruyt3

Department of Mechanical Engineering, University of Twente, 7500 AE Enschede, The Netherlandsn.p.kruyt@utwente.nl

1

Present address: FMC Technologies/CDS Separation Systems, Arnhem, The Netherlands.

2

Present address: Suzlon Blade Technology, Hengelo, The Netherlands.

3

Corresponding author.

J. Fluids Eng 132(6), 061104 (Jun 15, 2010) (8 pages) doi:10.1115/1.4001803 History: Received June 30, 2009; Revised April 28, 2010; Published June 15, 2010; Online June 15, 2010

Two-dimensional particle image velocimetry measurements and three-dimensional computational fluid dynamics (CFD) analyses have been performed on the steady velocity field inside the shrouded impeller of a low specific-speed centrifugal pump operating with a vaneless diffuser. Flow rates ranging from 80% to 120% of the design flow rate are considered in detail. It is observed from the velocity measurements that secondary flows occur. These flows result in the formation of regions of low velocity near the intersection of blade suction side and shroud. The extent of this jet-wake structure decreases with increasing flow rate. Velocity fields have also been computed from Reynolds-averaged Navier–Stokes equations with the Spalart–Allmaras turbulence model using a commercial CFD code. For the considered flow rates, the qualitative agreement between measured and computed velocity profiles is very good. Overall, the average relative difference between these velocity profiles is around 5%. Additional CFD computations have been performed to assess the influence of Reynolds number and the shape of the inlet velocity profile on the computed velocity fields. It is found that the influence of Reynolds number is mild. The shape of the inlet profile has only a weak effect at the shroud.

FIGURES IN THIS ARTICLE
<>
Copyright © 2010 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Meridional geometry and blade angle distribution of the impeller; the blade angle (in degrees) of the backward-curved blades is defined with respect to the circumferential direction

Grahic Jump Location
Figure 2

Schematic cross section of the experimental setup. The flow path is indicated by arrows.

Grahic Jump Location
Figure 3

Measurement planes in the impeller

Grahic Jump Location
Figure 4

Overview of impeller and PIV measurement area (left) and a typical PIV image (right)

Grahic Jump Location
Figure 5

Fine-grid solution with discretization error bars from the grid convergence analysis for the magnitude of the dimensionless relative velocity for Q=Qd; data along a circular arc with r/D=0.4 from blade pressure side to blade suction side; Ute=(ΩD)/2. The position along the circular arc is given by the dimensionless angle ξ (ξ=0 at the pressure side; ξ=1 at the suction side). Top, at hub; bottom, at shroud.

Grahic Jump Location
Figure 6

Contour plots of measured magnitude of the dimensionless velocity W/Ute for various flow rates; Ute=(ΩD)/2. Top, at hub; bottom, at shroud. The top left plot also shows two white circular arcs with radii r/D=0.3 and r/D=0.4. Impeller blades are shown in black.

Grahic Jump Location
Figure 7

Contour plots of computed magnitude of the dimensionless velocity W/Ute for various flow rates; Ute=(ΩD)/2. Top, at hub; bottom, at shroud. The top left plot also shows two white circular arcs with radii r/D=0.3 and r/D=0.4. Impeller blades are shown in black.

Grahic Jump Location
Figure 8

Comparison at hub and shroud of measured and computed profiles of the magnitude of the relative velocity vector along a circular arc at r/D=0.3 (top plots) and at r/D=0.4 (bottom plots); Ute=(ΩD)/2. The position along the circular arc is given by the dimensionless angle ξ (ξ=0 at the pressure side; ξ=1 at the suction side). The PIV measurements are indicated by symbols, while the CFD results are indicated by solid lines.

Grahic Jump Location
Figure 9

Influence of Reynolds number Re on the dimensionless velocity profiles for Q=Qd along a circular arc with r/D=0.4 from blade pressure side to blade suction side; Ute=(ΩD)/2. The position along the circular arc is given by the dimensionless angle ξ (ξ=0 at the pressure side; ξ=1 at the suction side). Top, at hub; bottom, at shroud.

Grahic Jump Location
Figure 10

Influence of the inlet velocity profile on the dimensionless velocity profiles for Q=Qd along a circular arc with r/D=0.4 from blade pressure side to blade suction side; Ute=(ΩD)/2. The position along the circular arc is given by the dimensionless angle ξ (ξ=0 at the pressure side; ξ=1 at the suction side). Top, at hub; bottom, at shroud.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In