TASK Quarterly   Scientific Bulletin of the Centre of Informatics - Tricity Academic Supercomputer & networK   ISSN 1428-6394

3-4/2008 XVIII Polish National Conference of Fluid Mechanics: selected CFD papers

Issue editors:
prof. dr hab. inz. Piotr Doerffer,
Institute of Fluid-Flow Machinery PAS, Gdansk, Poland
dr inz. Ryszard Szwaba,
Institute of Fluid-Flow Machinery PAS, Gdansk, Poland



  • D.Simurda, M.Luxa, P.Safarik, J.Synac, Aerodynamic Research on the Midsection of a Long Turbine Blade

    The paper is concerned with experimental aerodynamic research on the midsection of a 1220mm long turbine rotor blade. Optical as well as pneumatic measurements of the midsection blade cascade have been performed in a suction type high-speed wind tunnel. The results of measurements are analyzed and discussed.

    Interferograms and schlieren pictures taken in a wide range of isentropic exit Mach numbers and incidence angles exhibit the existence of several phenomena occurring in the transonic flow field at certain conditions concerning the exit Mach number and the angle of incidence. A flow separation taking place at an extreme negative incidence has been found to produce an additional loss of 6%. The presence of the reflection of an exit shock wave on the suction side of the neighbouring profile has been found to have a substantial influence on the losses, since the loss coefficient value has increased about 10% in cases without the reflection, i.e. the cases at a high exit Mach number and a high positive angle of incidence. Several reflection types have been observed and described.

  • J.Swirydczuk, Development of Vortex Structures in a Turbine Stage Rotor Passage

    The article analyses the formation and development of horseshoe vortices in a high-pressure turbine stage rotor passage. Two turbine stages are examined: a standard performance stage, revealing regular performance characteristics and distributions of flow parameters, and a low-efficiency stage in which a large separation zone is observed in rotor passages. In the latter stage the interaction of the hub horseshoe vortex with the separation structures has been found to take a highly unsteady and periodic course and be a source of remarkable flow fluctuations.

  • K.Urbanowicz, Z.Zarzycki, Transient Cavitating Pipe Flow: Computation Models and Methods

    The paper presents four key mathematical models of a transient cavitating pipe flow, i.e. the column separation model (CSM), the gas cavitation model (CSMG), Adamkowski's model (CSMA) and the bubbly cavitation model (BCM). All models investigated in the paper take into account unsteady frictional loss models. The equations describing all models have been solved using the method of characteristics at first and the finite differences method then. The results of numerical simulations have been compared with the results obtained in the experiments. Transients which have taken into account the unsteady wall shear stress fit well with the results of experiments in comparison with the quasi-steady wall shear stress model.

  • K.Bochon, W.Wroblewski, S.Dykas, Modelling of Influence of Turbulent Transition on Heat Transfer Conditions

    This article deals with the turbulent transition phenomenon modelling and its influence on heat transfer. The purpose of the analyses was to verify the transition modelling implemented in the ANSYS CFX 11 commercial code for popular test cases (low flow speed) described in literature, and then use it for verification of the in-house CFD code (created for compressible flows). The in-house CFD code has been extended lately for the Conjugate Heat Transfer modelling (CHT) as well, taking into account important flow effects, especially the turbulent transition. A Wilcox k-ω turbulence model with the Low-Reynolds modification was used in the in-house code. The calculations in ANSYS CFX were made using an SST turbulence model and a γ-Θ transition model. A fully turbulent flow was modelled by means of both codes, and the results were compared with the available experimental data. Then, the turbulent transition for several test cases was analysed with ANSYS CFX. Afterwards, the in-house CFD code was verified by means of ANSYS CFX for a higher flow speed (Mach numbers). The CHT modelling was analysed by means of both codes and the results were compared and discussed. The conducted analyses show that the results obtained by means of both codes are comparable, but the turbulence model used in the in-house CFD code is simpler and requires less computation time. A modification of two equations turbulence models can be an alternative for design problems in more developed laminar/turbulent flows.

  • M.Karczewski, J.Blaszczak, Performance of 3 Turbulence Models in 3D Flow Investigation for a 1.5-stage Turbine

    The object of this study was to investigate the flow phenomena in a cold air turbine built at the Institute of Jet Propulsion and Turbomachinery at Aachen Technical University (IST RWTH Aachen, Germany). The said turbine had been studied previously both experimentally and numerically on an IST's flow solver called Panta Rhei. Since that time certain improvements, computational-wise, have been implemented in the code. In order to test them, new simulation runs were conducted. The detailed studies of the measured and computed flow angles as well as a flow velocity analysis are the means for this evaluation.

  • M.Huptas, W.Elsner, Steady and Unsteady Simulation of Flow Structure of Two Surface-mounted Square Qbstacles

    The paper presents the numerical results obtained with the use of the FLUENT commercial code for analysing the flow structure around a single cube and two rectangular in-line surface-mounted bluff bodies immersed in a boundary layer. In the former case, clear effects of the inflow boundary layer thickness on the wall-shear stress within the wake of a single body are described. In the latter case, the grid resolution accuracy in predicting periodic vortex shedding from two tandem arrangement bodies is examined. Moreover, the aim of this study is to highlight the differences between steady and unsteady simulations.

  • M.Szymaniak, New Approach to Calculations of Turbine Stages and Regeneration Extraction Points

    The article presents 200MW LP turbine stage calculations taking into account leakage flows over rotor blades in the regeneration extraction point area. A methodology is described which allows the user to shorten the time-consuming CFD calculations solving the Navier-Stokes equation system in the examined area. A two-stage procedure was applied in which two types of calculations were coupled together. The first type is a one-to-one passage calculation of a steam flow through the turbine stages in the vicinity of the extraction point. This type of calculations preserves the circumferential periodicity condition. The second type is a circumferentially non-symmetrical calculation of the flow through an inter-stage diffuser with an extraction chamber. The calculations were preformed using 3D CFD codes, FlowER and FLUENT, in the two above mentioned areas, respectively. The solution was found using an iterative procedure for these areas coupled by boundary conditions, until convergence of calculations was reached.

  • M.Kabacinski, C.Lachowicz, J.Pospolita, Analysis of Flowing Fluid Effect on Flow Averaging Tube

    This paper presents a mathematical model and results of numerical simulations for a fluid flow around a flow averaging tube. The calculations have been performed using the commercially available FLUENT software. The authors have applied the currently known studies of the models of turbulence and their applicability in certain flow conditions and hence selected the RNG k-ε turbulence model including appropriate functions to be used for determination of pressures and velocities at the sites of occurrence of considerable gradients. The distributions of pressures and velocities around a sensor are presented along with pressure distributions, instantaneous and averaged in time, on the measuring tube surface. The paper determines the frequencies of measuring tube free vibrations for one sided and two-sided tube fitting for the tube length in the range 100-1500mm. This analysis has been conducted with the aid of solving equations for free undamped vibrations for specific models. The graphical presentation involves an admissible range of tube lengths with one and two-sided fitting for the specified flowing air velocities.

  • P.Doerffer, O.Szulc, Numerical Simulation of Model Helicopter Rotor in Hover

    The article presents details of a URANS simulation of the flow field near a hovering model of the Caradonna and Tung helicopter rotor (Caradonna F X and Tung C 1981 Experimental and Analytical Studies of a Model Helicopter Rotor in Hover, NASA Technical Memorandum). The CFD code SPARC (Magagnato F 1998 TASK Quart. 2 (2) 215-270) proves to be capable of capturing the aerodynamics of a two-bladed rotor in high-speed transonic hover conditions. A comparison of the simulation results with the experimental data is acceptable, hence the described methodology might be used with confidence in future numerical studies of application of noise-reducing devices on helicopter blades.

  • P.Mirek, J.Jablonski, W.Nowak, Numerical Optimization of Air Flow in the Plenum Chamber of an Industrial CFB Boiler

    The article presents the results of experimental and numerical studies on windboxes operating in a 235MWe Circulating Fluidized Bed (CFB) boiler. The main problems of windbox designs have been identified and a modified internal geometry has been proposed which causes a more uniform flow under the grid and prevents the velocity field from formation of dead zones.

  • S.Dykas, A.Wilk, Determination of the Flow Characteristic of a High-Rotational Centrifugal Pump by Means of CFD Methods

    The paper refers to a numerical analysis of the flow through a centrifugal pump working at high rotational speed. This kind of a pump is characterized by a high delivery head and a low discharge. The considered pump is a one-stage centrifugal pump with a rotational speed of 12000rpm. This pump has been manufactured and prepared for experimental research. The main purpose of the presented research has been to determine the flow characteristic of the pump by means of CFD methods. To this end the commercial CFD code, ANSYS CFX 11, has been used. The obtained numerical results have shown high usefulness of the CFD methods for the flow machinery design process. In the next step the obtained flow characteristic will be compared with the experimental investigation.

  • M.Marek, W.Aniszewski, A.Boguslawski, Simplified Volume of Fluid Method (SVOF) for Two-Phase Flows

    The PLIC approach has been usually used in recent implementations of the VOF method i.e. the interphasal surface is approximated by a plane with an arbitrary orientation with respect to the computational cell. Although this method is accurate, it is rather difficult to implement, as a large number of orientations need to be taken into account and the calculation of volume fraction fluxes is not straightforward. A simpler approach to VOF - SLIC - requires much less effort from the programmer but the interface approximation by a plane parallel to the cell surfaces is too crude and the results are not satisfactory. The method presented in the article may be considered as an intermediate approach between PLIC and SLIC - fluxes are computed directly only for the interface's special orientations and linear interpolation is used for calculation of the fluxes for the remaining cases. Some classical tests of the proposed method are performed and an example of a broken dam problem simulation is presented.

  • W.Majchrowski, E.Tuliszka-Sznitko, Rotating Rayleigh-Benard Convection

    In the this paper our results on the natural convection in an enclosed rotating cavity are presented. We have focused our attention on the influence of the Rayleigh and Taylor numbers on the flow structure. DNS computations have been performed for the geometry of aspect ratio L = 9 and curvature parameter Rm = 1.5.

  • Z.Kosma, Fast Algorithms for Calculations of Viscous Incompressible Flows Using the Artificial Compressibility Method

    An artificial compressibility method is designed to simulate stationary two-and three-dimensional motions of a viscous incompressible fluid. A standard method of lines approach is applied in this contribution. A partial differential equation system is discretized in space by second-order finite-difference schemes on uniform computational grids, and the time-variable is preserved as continuous. Initial value problems for systems of ordinary differential equations for pressure and velocity components are computed using the Galerkin-Runge-Kutta method of third order. Some test calculations for laminar flows in square, cubic, triangular and semicircular cavities with one uniform moving wall and double bent channels are reported.

  • Z.Kosma, P.Motyl, Fast Algorithms For Calculations of Viscous Incompressible Flows Using the Velocity Correction Method

    The velocity correction method is designed to simulate stationary and non-stationary, two- and three-dimensional motions of a viscous incompressible fluid. The basic assumption of this method consists in splitting the velocity and the pressure fields and calculations are performed in two steps. In the first step, a tentative velocity field is determined by simplified equations for momentum conservation. In the second step the Neumann problem for the Poisson equations is solved to obtain the computational pressure, and the velocity components are corrected. A standard method of lines approach and the two grids method are applied in this contribution. Some test calculations for laminar and transitional flows in square and cubic cavities with one moving wall as well as in a backward-facing step are reported.

  • Z.Wiercinski, A.Skotnicka-Siepsiak, Application of CFD for Temperature and Air Velocity Distribution Calculation in a Ventilated Room

    Appropriate air distribution in a room is necessary for thermal comfort. By using Computational Fluid Dynamics (CFD) it is possible to compare optional ways of air supply and distribution at the stage of the ventilation design concept. Using these simulations the ventilation system designer can choose the best method of air supply in the room diminishing the risk of an incorrect solution.