Design of an impeller for an high speed cavitation tunnel

Cavitation tunnels are usually used to measure the performance of marine propellers, formation of cavitation, noise characteristics, etc. and they work as closed loop systems. On the otherhand; they can be used for various tests of model torpedos, hydrofoils or plates, depending on the general dimension of test section and the velocity of flow. Velocity of the water in a cavitation tunnel is controlled by an axial impeller which is located in the bottom part of the tunnel. Main source for the noise within the cavitation tunnels could be flow separation, high turbulence level, cavitation, etc. generated by the impeller which circulates the volume of flow. In order to obtain a good cavitation tunnel characteristics an appropriate impeller design is one of the key issues by taking into account all the above factors. Within the scope of the project called "Cavitation Tunnel and Manoevring Test Systems for Improvements of Hydroacoustic Properties of Turkish Navy Ships, a large-scale cavitation tunnel will be built in the Faculty of Naval Architecture and Ocean Engineering of Istanbul Technical University. At the outset of the study presented in this master's thesis, the aim of the study was specified to design an impeller for the cavitation tunnel, as being one of the crucial steps for the cavitation tunnel to be built. Chapter 1 reviews the state-of-the-art literature studies regarding cavitation tunnels and impeller design. In Chapter 2, the methods used to calculate the energy loss is discussed in details, which is an input power of the axial impeller. Additionaly, according to the tunnel criteria, hydrodynamic design steps of axial type impeller were explained. Hydrodynamically designed axial impeller, have been examined by the use of Computational Fluid Dynamics (CFD) method, Reynolds Averaged Navier Stokes (RANS). RANS (Reynolds Averaged Navier Stokes) equations are time averaged equations for incompressible fluid flow. Reynolds decomposition is the main idea of the equation, whereby an instantaneous quantity is decomposed into its time-averaged and fluctuating quantities. Generating a good mesh configuration is very important to get reliable results. After having prepared a suitably meshed model, the analyse can be run. Required reports can be written during the analysis, reports can be monitored for each iteration step. Residuals can be monitored to have opinion if the solution converges or not. In this study, resistance forces were reported and monitored during the analyses. Stopping criteria is a point that should be mentioned. In order to make less iteration, a stopping criteria can be defined. When the specific number of iteration value converge until the criterion has satisfied, the analysis stops. The detailed information about the methodology of the CFD analysis has been presented in Chapter 3. Moreover, the finite volume method, which was used in the flow analysis and SIMPLE method, which is used to separate speed and pressure were explained along with the SST k-ω model, which was used as the turbulence model. ANSYS Fluent 16.2 was used as the cell centered finite volume solver. Chapter 4 presents numerical studies carried out that are the applications of the methods described in Chapter 2 and Chapter 3. In this chapter, the energy loss in the cavitation tunnel sections were calculated. Geometry of the axial impeller were formed and computational studies were performed. In addition to the results, the profile types used for the section of the impeller, modelling of impeller blade surfaces, flow domain and solution mesh, mesh dependency and validation studies were also given. Geometry of an axial impeller have been decided by taking into consideration the fluid properties, flow rate, efficiency, input power, head of pump, rotational speed, specific speed, specific energy, number of blades and other design criterias. The total energy losses calculated, was evaluated together with the mechanical and hydraulic efficiency characteristics. As a result of all these steps, the engine power needed to satisfy the maximum flow velocity of 15 m/s was calculated. Power requirement of the tunnel was calculated according to energy loss for 2, 10 and 15 m/s velocity cases. Moreover, the hydrodynamic design of the axial impeller was performed based on the calculated energy loss for 15 m/s velocity. General dimensions of the cavitation tunnel and the diameter of the section in which the impeller will be placed, are known. The diameter of the impeller was decided according to the clearance between the blade tip of the propeller and the surface of the duct. The type of profiles, which were used for the impeller blade sections, has been decided by parametric use of the angle of attack based on the calculated lift coefficient values were obtained experimentally for a single profile (2D) and the profile located in the cascade (3D) cases. Difference between these values were also calculated. The angle of attack was varied until to mimimise difference between the lift coefficients. The two lift coeffcients were calculated, using empirical methods. Once the angle of attack has been determined, 4 digit NACA series were selected for the sections. The lift coefficients of the selected profiles were also compared by the results obtained in the 2D CFD analysis. In the computational studies, the same angle of attack values, which were obtained by the empirical methods, were used. A total of 6 different section profiles was selected based on NACA 4 digit series while generating the geometry of the impeller blade. In these profiles, from the hub to the tip of blade, the angle of attack values of -0.40, 1.75, 1.65, 1.20, 0.85 and 0.65 were used respectively. The mean line of profiles were formed according to the design criterion. 4 digit NACA series of 9406, 9412, 9410, 6407, 4405 and 3403 from the hub up to the tip of blade were used, respectively to satisfy the camber values obtained previously. These profiles were decided by considering the position of the maximum camber and thickness. The hydrodynamic design of the axial impeller in 3-Dimensional form has been completed by using the profile types and lean angles, which were obtained according to design criterion and the total energy loss that was calculated based on the geometry of the sections. The impeller was further analysed by using the RANS method and the solution mesh structure of the impeller was generated based on the findings obtained in the mesh dependency and validation study. As a result of the computational studies, the rate of revolution of the impeller has been optimised and the the total energy loss, hence the power values required to circulate the flow were calculated for each velocity condition, ranging from 2 m/sn to 15 m/sn.