Archive for December 2013

Mesh Quallity


Mesh quality


For the same cell count, hexahedral meshes will give more
Accurate solutions, especially if the grid lines are aligned with the
Flow.
The mesh density should be high enough to capture all relevant
Flow features.
The mesh adjacent to the wall should be fine enough to resolve
The boundary layer flow. In boundary layers, quad, hex, and
Prism/wedge cells are preferred over tri’s, tets, or pyramids.
Three measures of quality:
Skewness.
Smoothness (change in size).
Aspect ratio.

Skewness

Two methods to define skewness are:
·        Based on equilateral volume :

o   Skewness = (optimal cell size – actual cell size) / (optimal cell size)


o   Applied to only tri and tet meshing.
o   This method on meshing follows Delaunay triangulation theorem.
o   Default method used in ansys for tri and tet meshing.



·        Based on the deviation from a normalized equilateral angle
o   Skewness = Max [ {(θmax -90)/ 90}, {(90 – θmin )/90}]


o   Applies to all cell and face shapes, used generally for quads.
o   Used for prisms, pyramids, etc.

Smoothness and aspect ratio
  • Change is mesh size should be gradual or smooth


  • Aspect ratio is the ratio of longest edge length to the shortest edge length. It is equal to 1 for an equilateral triangle or square.

Checking For Quality



  • A poor quality grid will cause inaccurate solutions and/or slow convergence.
  • Minimize equiangle skew.
  • Hex and quad cells: skewness should not exceed 0.85.
  • Tri’s: skewness should not exceed 0.85.
  • Tets: skewness should not exceed 0.9.
  • Minimize local variations in cell size.
  • If such violations exist: delete mesh, perform necessary decomposition           and/or pre-mesh edges and faces ,and remesh.
  • More cells can give higher accuracy. The downside is increased memory and      CPU time.   
  • Use a non-uniform grid to cluster cells only where they are needed. For eg        while doing airfoil analysis accumulate the grid at the leading and trailing        edges and away from the boundary





  • ·       




    Tuesday 31 December 2013
    Posted by Unknown

    Solver Study: Turbulent Flow Models


    Turbulence

    Turbulence or turbulent flow is a flow characterised by random and chaotic fluid flow with unexpected property changes.



    Turbulence Models

    A turbulence modelling is a computational procedure to close the system of mean flow equations. For most engineering applications it is unnecessary to resolve the details of turbulent fluctuations. Turbulence models allow the calculation of the mean flow without first calculating the full time-dependent flow field. we only need to know how turbulence affected the mean flow.

    The classical turbulence models are based on the Reynolds Averaged Navier-Stokes equations (RANS) :


    One equation model: Spalart-Allmaras
    1. Two equation model: -> k-epsilon (Standard, RNG and realizable)
                                    -> k-omega
    2. Seven equation: Reynolds stress model (RSM)
    Spalart-Allmaras one equation model:



    This model solves a single conservation equation (PDE) for the turbulent viscosity. It contains both connective and diffusive transport terms as well as expressions for the productive dissipation of turbulent viscosity. 

    It was developed for the use in unstructured codes in the aerospace industry. It gives accurate results for attached wall-bounded flows and flows with mild separation and recirculation.  however it gives improper results for massively separated flows, free shear flows and decaying turbulence. 

    Can be used for:

    • Drag prediction with mild turbulence
    • Wing parameter calculation for mild separation conditions
    K epsilon 2 equation model:



    The transport variable used in the equations is the turbulent kinetic energy. The second transported variable in this case is the turbulent dissipation (epsilon). The K-epsilon model has been shown to be useful for free-shear layer flows with relatively small pressure gradients. Similarly, for wall-bounded and internal flows, the model gives good results only in cases where mean pressure gradients are small; accuracy has been shown experimentally to be reduced for flows containing large adverse pressure gradients. It has 3 variations : 
    1. Standard k epsilon model: Good for simple simulations
    2. Realisable k epsilon model : Improved performance for flows involving boundary layers under strong adverse pressure gradients or separation, rotation, re circulation and strong streamline curvature
    3. RNG k epsilon model : It offers improved accuracy in rotating flows,  favoured for indoor air simulations. better results for rotating flow and effect of swirl on turbulence
    It is a really simple model to implement and leads to stable calculations that converge easily and gives reasonable predictions for many flows. 

    although it is not favored for rotating and swirling flows.
     
    k omega 2 equation model


    It is another 2 equation model similar to k epsilon model which solves 2 additional PDE's resulting in a faster converging solution. however it is not widely used because it might give errors(An assumption based on trial and error) 

    Reynolds stress seven equation model


    RSM closes the Reynolds-Averaged Navier-Stokes equations by
    solving additional transport equations for the six independent
    Reynolds stresses. It is a good model for accurately predicting complex flows. Accounts for streamline curvature, swirl, rotation and high strain rates.

    The rate of convergence is low but the results are accurate and it can handle complex flows well

    Wall treatment in turbulent models

    A wall treatment is the set of near-wall modelling assumptions for each turbulence model. The wall functions are a set of semi empirical functions used to satisfy the physics of the flow in the near wall region. Turbulence is affected in many ways by the presence of the wall through the non slip condition that must be satisfied at the wall. Four areas in the near wall region are defined, the laminar sub-layer, the blending region, the log law region and the outer region. Each region has a different effect on turbulence. In ansys Fluent they are of 4 types :

    1. Standard wall functions : A wall treatment is the set of near-wall modelling assumptions for each turbulence model.
    2. Non-equilibrium wall function : When the near-wall flows are subjected to severe pressure gradients, and when the flows are in strong non-equilibrium, which means that the turbulence production term and the dissipation term are not equals, the results given by the standard functions are not satisfactory enough and the non equilibrium wall function allows for better calculations.
    3. Enhanced Wall Functions The enhanced function in FLUENT are used to achieve near wall modelling approach having the accuracy of the standard two layer approach for fine meshes and at the same time not degrading the results for the wall function meshes. In order to do so the enhanced wall functions are combined with the two layer model.


    Thursday 26 December 2013
    Posted by Unknown

    Solver Study: Inviscid and Laminar Models

    Inviscid Flow



    An inviscid flow is the flow of an ideal fluid that is asumed to have no viscosity. In fluid dynamics there are problems that are easily solved by assuming that the flow is inviscid. It is a flow in which the viscous effects can be neglected.

    The flow of fluids with low values of viscosity agree closely with inviscid flow everywhere except close to the fluid boundary where the boundary layers plays a significant role. At high reynolds numbers, flow past slender bodies involve thin boundary layers. Viscous effects are important only inside the boundary layer and the flow outside it is nearly inviscid. If the boundary layer is not separated then the inviscid flow model can be used to predict the pressure distribution with reasonable accuracy. 

    The inviscid flow model in ansys fluent can be used for simulations where nature of flow in not important to the problem like all the simple heat flow problems. Although air flow on an aircrafts body in with very high reynolds numbers(nearly inviscid) but the assumption of completely ignoring viscosity result in poor results and create problems while analysing high lift devices or analysis at high AOA. For better results the laminar flow model should be used.


    Laminar Flow



    Laminar flow is a type of fluid flow in which the fluid travels on smooth or in regular paths, in contrast to turbulent flow, in which the fluid undergoes irregular fluctuations and mixing. In laminar flow, sometimes known as streamline floe, the velocity, pressure and other flow properties at each point in the fluid remain constant. Laminar flow over a horizontal surface may be thought of as consisting of thin layers, or laminae, all parallel to each other, The fluid in contact with the surface is stationary, but all the layers slide over each other.  

    For eg. Laminar flow in a straight pipe may be considered as the relative motion of a set of concentric cylinders of fluid, the outside one fixed at the pipe wall and the others moving at increasing speeds as the centre of the pipe is approached. Smoke rising in a straight path from a cigarette is undergoing laminar flow. After rising a small distance, the smoke usually changes to turbulent flow , as it eddies and swirls from its regular path





    Wednesday 25 December 2013
    Posted by Unknown

    Popular Post

    Powered by Blogger.

    - Copyright © Ankit Saxena Powered by Blogger - Designed by Shachi Vaish