Learning Forum

[Alexia16]

Hi, I am doing a project on simulating a wing with its winglet, and the meshing is resulting in sharp leading edge curvature as shown in the picture below:

[Keyur Kanade]

This could be due to geometry problem or due to wrong mesh sizing. Can you please check following? nPlease check geometry in SpaceClaim. Select geometry in structure tree --> Use right click --> Select Check Geometry. The geometry should be error free to proceed. If geometry has any errors, please modify/recreate geometry at those places.nIf geometry is error free, then please select 'CFD' as Physics Preference in meshing. nPlease give body/face/edge sizing so that mesh has sufficient mesh density and resolve important features. nFor the error, please use right click on the error and click on 'problematic geometry'. This will highlight geometry which is failing to mesh. You will need to modify geometry at this location or change mesh sizing. nIf your meshing has special controls like periodicity, please check if those are correctly defined. nAlso try to split the wing face into 3-4 faces in SpaceClaim. nRegards,nKeyurnHow to access Ansys Online Help DocumentnHow to show full resolution imagenGuidelines on the Student CommunitynHow to use Google to search within Ansys Student Communitynn

[Alexia16]

Thanks for replying.nI followed your advice and checked the geometry in space claim, but no error was found. The physics preference has been set to CFD as follows:nI have added local control over the sizing of the body of influence, wing face, and wing leading edge. The mesh does not make use of any periodic boundary. There is no error pop up with reference to any geometry, only inflation creating stairstep error as follows:nI forgot to add the following in the original post:nThe leading edge of the wing mesh results as follows:nThe number of elements: 24805549 Nodes: 4338225 n

[Alexia16]

n

[Rob]

Have a look at the curvature sizing function, that usually fixes this sort of issue. n

[Alexia16]

Thanks for replying. I tried using the curvature sizing function, as well as adding edge sizing to the wing cross-sectional edges, which resolved the issue of sharp leading-edge, however, it gave me around 22 million elements and I don't have enough computational power to run the solution. Increasing the sizing functions is giving me the ' Floating Point Exception' error when I try to run the simulation. I tried another mesh combination (does not include inflation) which gave me the following mesh:nnHowever, the face mesh in the tree outline became as follows:nwhich could be due to improper selection for the face mesh, however, I don't understand why in this case, the domain mesh near the leading edge of the wing followed the airfoil shape precisely, while when the 'face mesh' is successful (green check), it again results as following:nFor some reason, my mesh is not generating at all without adding the 'Face Mesh' function, unless for the two cases as I described above. Anyway to solve this issue please?n

[Rob]

Looking at the shape is it an extruded 2d object? If so you can run in 2d mode, that'll significantly reduce the cell count. Or use sweep: you need to read up on inflation & sweep as it's a little different to the unstructured inflation approach. Then look at the bulk domain cell size: you want to grow away from the wing to avoid wasting cells but also retain enough mesh to pick up any separation & wake regions. n