CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   meshing lubricant volume (http://www.cfd-online.com/Forums/fluent/27812-meshing-lubricant-volume.html)

 kenneth August 2, 2000 00:43

meshing lubricant volume

hi,

i am using gambit to create an 3d journal bearing and encounter problem of meshing the lubricant volume. it is a very thin film between the bearing and journal and needs to be meshing for fluent 5 at later application.I am unable to mesh the lubricant volume succeesfully.

is there any advices for me to solve this problem??

regards

kenneth

 John C. Chien August 2, 2000 12:08

Re: meshing lubricant volume

(1). Two solutions,: (a). use other programs, (b). use copying machine to enlarge the geometry several times, until you can put some 20 lines between the surfaces. The structured hex grid is ideal for this application. (2). Everything is relative, so, you need to resolve the film thickness first by enlarging the geometry. A minimum of 10 points across is needed to produce a good laminar velocity profile.

 kenneth August 3, 2000 00:41

Re: meshing lubricant volume

hi john,

thanks for your advice..are u refering to enlarge the geometry using gambit software or other means and mesh the volume of lubricant by interval count of 10...on the surface of lubricant.

regards kenneth

 John C. Chien August 3, 2000 02:46

Re: meshing lubricant volume

(1). It does not matter which software you use. (2). The point I was trying to make is you need to isolate the thin layer region from the rest of region. Then you can put in adequate mesh density in the region. (3). There will be transition from the thin region to other thicker region, but you can fallow the same principle and isolate the transition region from the rest of the flow field. Then create a mesh for this transition zone. (4). And step by step, you will be able to cover the whole flow field. Any mesh generation and geometry package which can handle bottom up approach (from point,vertex, line, curves, surfaces up to create the geometry and mesh), can handle the complex geometry problem by dividing the flow field into smaller blocks based on the characteristics of the geometry. (5). I have used the preBFC and ICEM/hexa, and think these are good and hard to use codes ideal for complex structured geometry.(a geometry which can be covered by multi-block mesh)

 Kai Kang August 3, 2000 09:34

Re: meshing lubricant volume

The reason being the thickness of the thin film is so small compare to the other dimensions in your model. You need to separate(split) this layer from the model and mesh it separately using the mesh density you want. If you are using tet mesh, this small thickness will probably give you a "gap" problem for the meshing. Anyways, try to avoid too large edges/too small edges in a single block...

 kenneth August 7, 2000 01:33

Re: meshing lubricant volume

hi guys,

sorry to bother u all again.

i have tried to apply your advice of meshing the lubricant volume and no success meshing in doing so. it gives me the error." no logical cube for meshing was to be found ". i use the following details to mesh the volume of lubricant...element :HEX, Type: Map ,interval count : 20.By the way, do i need to specified the boundary layer for the face and the edge of the lubricant for the meshing. hope to hear from u all soon.

regards kenneth

 Shyam Kishor August 7, 2000 12:37

Re: meshing lubricant volume

The error message suggests, you are trying to MAP mesh a volume (domain) which is not mapable. You need to decompose the volume (domain) into simpler ones, so that they become mapable, or use other mesh schemes. Also, use of boundary layers and/or edge meshing (grading) will help you achieve desired mesh density and quality.