Swirling flow in a conical diffuser generated with rotor-stator interaction

Application Challenge AC6-14 © copyright ERCOFTAC 2024

Overview of Test

A closed loop experimental test rig was developed and measuresd at the Politehnica University of Timişoara, Romania, [5] for investigating decelerated swirling flows in a conical diffuser and for assessing various flow control methods. The test rig includes the following (see Fig. 2): (i) a main hydraulic circuit, marked blue; (ii) an auxiliary hydraulic circuit, marked red that is implemented in order to assess different control methods [6, 3]; (iii) a lower reservoir with a volume of 4m³ (iv) a main pump that is able to provide a maximum flow rate of 40 l/s; (v) an upper reservoir equipped with a honeycomb section to provide an uniform flow at the inlet of the swirl test section, and (vi) a test section, marked magenta.


Figure 2: Photo and schematic view of experimental closed loop test rig from UPT

The test section contains two main parts: i) the swirl generator with 13 fixed guide vanes and a runner with 10 blades, see Fig. 3, and ii) the convergent-divergent test section [17]. The divergent part of the test section, with total angle of 17° (2 × 8.5° ) is designed to generate a flow similar to that in a Francis turbine draft tube. The hub and shroud diameters of the swirl generator are ${D_{hub}}$ = 0.09m and ${D_{shroud}}$ = 0.15m, respectively. The guide vanes create a tangential velocity component, yielding a pressure increase from the hub to shroud at all operating conditions. The purpose of the runner is to redistribute the total pressure by inducing an excess in the axial velocity near the shroud and a corresponding deficit near the hub, like a Francis turbine operating at partial discharge. The runner thus acts as a turbine near the hub and as a pump near the shroud. There is a gap between the blade tip and the shroud of about 0.4mm. Although the effect of tip-clearance is not investigated, the velocity profiles measured further downstream did not reveal significant leakage flows effects. The numerical simulations without and with the blade tip gap further support this conclusion.

All experimental data are gathered in single phase (non-cavitating) conditions by keeping the test rig pressurized at a large enough pressure level, since a cavitating vortex rope brings additional complexity to the flow phenomenon and the experimental methodology. The experimental velocity is obtained with a Dantec Dynamics two-component LDA. The LDA system consists of an argon-ion laser of 300mW power, and an optical probe with focal length 159.6mm, beam diameter 2.2mm and beam spacing 39.2mm. A three-dimensional traversing system is used for the probe positioning, with a 0.01mm accuracy on each axis. The measurements are performed along each survey axis with a step size of 1mm. Silver coated hollow glass particles are added into the water for efficient light back-scattering. The average particle diameter is 10μm, and the relative density is ${\rho _{particles}/\rho _{water}}$ = 1.402. For each experimental point there are between 20000 and 50000 particles crossing the measuring volume, during the acquisition time of 30 seconds.

Figure3: Photo of the Timişoara swirl generator

TEST CASE

Description of Experiment

The velocity measurements are performed along survey axes W0–W2 in the test section, shown in Fig. 4. The survey axes are perpendicular to the shroud. Survey axis W0 is 70mm downstream the inlet of the test section. Survey axes W1 and W2 are located in the divergent part of the test section 113mm and 168mm downstream of the inlet of the test section, respectively. The LDA measurements start measuring on the point at the wall of the window and move the probe along the survey axis to the opposite wall.

|

Figure 4: Photo and schematic view of the test section with survey axes for LDV measurements

Boundary Data

An upper reservoir equipped with a honeycomb section is employed to provide a uniform flow at the inlet of the swirl apparatus, as shown in Fig. 2. A lower reservoir is used with a volume of 4m³. The runner speed is 920rpm at a discharge of 30 l/s.

Measured Data

The survey axes, ${\textit {S*}}$ at sections W0-W2 (see Fig. 4b, are normalized by the throat radius, ${R_{throat}}$ = 0.05m, and the velocity is normalized by the bulk velocity, ${W_{throat}}$ = 3.81m/s, at the throat. The survey axes have their origin at the wall and are perpendicular to the wall. The mean meridian velocity component is measured along the survey axes. The meridian velocity is defined as the velocity component which is perpendicular to the survey axes. The meridian velocity is called axial velocity hereafter for simplicity. The mean tangential velocity component is measured along the survey axes. The length is based on the distance from the wall. The velocity fluctuations root mean square are included.