In the early days of aerospace turbomachines, the value of curved vanes for higher efficiency was appreciated while designing aeroengine superchargers during WWI1. Consequently, the first design trade-off was between mechanical stress (i.e., the impeller diameter) and efficiency. In particular, a straight vane produces a higher pressure head compared to a backward curved vane. Furthermore, as the impeller RPM and diameter (i.e., the impeller tip speed) increase, the straight vane produces lower bending stresses vis-à-vis the curved vane that tries to “straighten” itself.

A radial compressor can be divided into different parts, as shown in Figure 1.1. The flow is aspirated from the inlet plenum and after being deflected by the inlet guide vanes (IGV), it enters the inducer . From there on the flow is decelerated and turned into the axial and radial directions before leaving the impeller in the exducer . The presence of a radial velocity component is responsible for Coriolis forces, which, together with the blade curvature effect, tends to stabilize the boundary layer at the shroud and suction side of the inducer (Johnston 1974; Koyama et al. 1978). The boundary layer becomes less turbulent and will more easily separate under the influence of an adverse pressure gradient.

The flow in radial impellers is 3D, viscous, and unsteady. Unsteady 3D Navier–Stokes calculations require a very detailed geometry definition and considerable effort. The first Navier–Stokes steady flow calculations by Wallitt (1980) and Moore et al. (1983) nicely described the major flow phenomena such as secondary flows and suction side flow separation (Figure 3.1). However, excessive computational effort (6 weeks on a supercomputer for only 100 000 grid points in 1980) restricted their use to a final check of performance. The enormous increase in computational capacity and speed, and the improvement in the solvers, now allow a routine analysis of the 3D flow in impellers at design and off-design operation once the detailed geometry is defined.

This chapter provides a short overview of the different non-axisymmetric inlet and outlet geometries for radial impellers. It describes the advantages and disadvantages of the different types, and explains the relation between the geometry and the flow in terms of performance and circumferential pressure distortion. Inlet and outlet volutes are discussed separately because of the fundamental differences between the flows.

The flow perturbation by the diffuser vanes, a volute operating at off-design, and diffuser rotating stall are the main sources of the circumferential pressure distortion downstream of a centrifugal impeller. The direct consequences are the time varying flow and forces in the rotating blade passages (relative frame of reference) and a circumferential variation of the velocity at the diffuser inlet (absolute frame of reference). This chapter concentrates on the unsteady impeller response to the downstream pressure distortion, imposed by volutes operating at off-design. After a short overview of some experimental data, the following topics will be discussed:

The operational principle of a centrifugal fan (aka a radial turbo fan) accords with that of a centrifugal pump, but the fluid is different, now it’s a gas. It is built up out of an impeller with vanes that rotates fast in an abutting volute (Figures 9.1–9.4).The rotating movement of the impeller produces a pressure rise on the locked-in gas; the gas is propelled out of the impeller and caught in the volute. An underpressure arises in the inlet and this cause the gas to be sucked into the suction line.

The goal of this article is to provide an overview of basic principles that are important for design and evaluation of bearings used in blood pumps. While the primary focus is on mechanical bearings used in the second and third generation ventricular assist devices (VADs), bearings used in the first generation pulsatile devices as well as those used in percutaneous and extracorporeal applications are also described. The new generation of continuous flow VADs uses rotating impellers to accelerate the blood and increase the fluid pressure while delivering the required flow rate. These rotary pumps are classified as either axial or centrifugal flow devices. In the axial flow pumps, the fluid, in this case blood, enters at one end of the cylindrical pump housing, follows the axial direction of the housing, and exits at the other end. In contrast with this flow geometry, the impellers used in centrifugal pumps push the fluid out of the housing in the radial direction. In either case, the rotating impeller requires load support in the axial and the radial directions. This support mechanism is provided either by magnetic bearings or by blood-lubricated mechanical bearings.

As indicated in Figure 1-3, there are two types of bearing configurations: (1) thrust bearing to support the normal load and (2) journal bearing to support the radial load. Often separate thrust and journal bearings are designed and implemented to support a rotating impeller. In certain cases more complex designs are used to combine the two bearing types. Nevertheless, the basic geometry that makes load support possible in hydrodynamic bearings is a converging wedge. In other words, the fluid is pulled into a converging gap through shear and a force perpendicular to the flow direction is produced, providing load support. Analysis of flow and prediction of load carrying ability of bearings starts with the Reynolds equation.

The three-dimensional model of the impeller of High-pressure Small-Flow Centrifugal Pump has been established by Pro-E, and the structure strength and modal analysis are carried on impeller by ANSYS. The strain contour diagram, stress contour diagram, and vibration shape diagram of each step frequency in design condition are obtained. The results show that the strength design of the impeller accord with the design requirement, and the frequency of load will be chosen according to the results of modal analysis in order to avoid resonance.

Many of the dynamic analyses conducted during the design phase of a pump station or for trouble shooting problems are invariably involved in the startup and shutdown of centrifugal pumps, flow controls, and in the opening and closing of valves during operation scenarios. Interaction between system hydraulic dynamics and associated control protocols could also trigger severe dynamics and catastrophic failure or system overall reliability issues. Expansion of pump stations to cope with either capacity increases or changes in conditions or products can be a challenge. The addition of various units in series or parallel with different pump head-flow characteristics, and drivers' type and inertias, require attention and thorough analysis. In order to understand the dynamic behavior of a pumping system, the governing equation describing the unsteady incompressible flow through various piping elements needs to be solved in both temporal and spatial domains. A constant area pipe is described by the continuity and momentum equations. Therefore, the fundamental governing one-dimensional equation is described first and are solved using the method of characteristics. Since the pump driver exerts a torque on the rotating shaft, which in turn, conveys energy to the impeller to develop a total dynamic head across the pump, the dynamic behavior of this system is also described and combined with the unsteady flow of the connecting piping system. The transient-state pressures at some locations in the system may be reduced to or below the vapor pressure of the liquid. This may produce vapor cavities in the flow or may cause the liquid column to separate. Rejoining of the separated columns or collapse of the cavities results in a large pressure rise, which may damage the piping system. This chapter deals with these dynamic phenomena frequently observed in pumping systems.

The total enthalpy rise imparted to the fluid by the impeller is termed the impeller work input. An accurate prediction of the impeller work input is fundamental to all aspects of centrifugal compressor aerodynamic design and analysis. As discussed in Section 1.3, the work input supplied by the impeller blades is the useful work, which supplies the pressure rise. There are also various sources of parasitic work, which can be considered wasted energy. The fluid in the clearance gaps between the impeller and the housing exert frictional forces on the impeller, consuming part of the power supplied to it. This is a parasitic work commonly called the windage and disk friction loss. For covered impellers, fluid leakage from the impeller discharge back to the inlet through the impeller eye seal (see Fig. 1-4) is another source of parasitic work, known as the leakage loss. Since the pressure rise imparted to this fluid is dissipated in the clearance gap and seal, the impeller works on this portion of the fluid twice, with no addition contribution to the pressure rise. These two sources of parasitic work are easily recognized physical phenomena. From experience, it is known that other parasitic work models are needed to explain the observed work input curves from compressor test results. These take the form of models postulated from fluid dynamics reasoning and are validated by comparing predictions with experimental data. Here, the models of Aungier (1993b, 1995) will be reviewed. It is reasonable to also expect a leakage loss in open impellers, where the fluid leaks through the blade clearance gaps, dissipating a portion of the pressure rise, to be re-energized by the impeller in the next blade passage. In highly diffusing impellers, another parasitic work, the recirculation loss, is observed. It is believed to result from a portion of the fluid at the impeller tip reversing to re-enter the impeller at the tip, again dissipating part of its pressure rise and requiring the impeller to re-energize it. It is convenient to express the impeller work in dimensionless form as the work input coefficient defined in Eq. (1-6).

The term “gas path design” refers to the specification of the geometry of a component’s surfaces that bound the passages through which fluid flows. This includes the hub-and-shroud contours and the blade surfaces for vaned components. This chapter describes some fairly general procedures for gas path design and their application to detailed impeller design. These gas path design procedures can be applied to the design of other stage components, for which very general design methods are needed. As will be seen in the following chapters, the detailed design of most centrifugal compressor components can be directly accomplished with methods specific to those components. Hence, the most common application of the methods of this chapter is impeller detailed design. Very little additional effort is required to implement the procedures in this chapter in a form applicable to any component consisting of an annular passage, with or without vanes, which covers most of the stage components of interest. On occasion, a more general design than that offered by the component-specific methods in the following chapters is needed. For example, there can be a need for three-dimensional vane geometry for vaned diffusers or even return channels, which is not offered in the procedures presented in Chapters 9 and 10. If implemented properly, the methods of this chapter will provide that more general, detailed design capability when required.