[Светлана]

Hi

I have a natural convection problem near a plate there is many little detail and I am not sure how to make meshing possible. I have created a test case, it is described below.

I have a flat plate. At each 4 square centimeters there is one box sitting on the plate. Think of it like someone put many lego bricks on the floor evenly with each lego brick being 2cm wide and 2cm deep and 2cm tall, with 3cm gap between these blocks. I am interested in the fluid flow along this plate with boxes on it. The plate is 1m x 1m size. It is positioned vertically and is in a constant temperature of 50 degrees C with 20 degrees C ambient. The fluid is air.

I've added the geometry sample at http://sveta.uk.to:2018/~svetlana/CFD/test.wbpz .

In ansys mesh I set the following:

- face sizing on the boxes and on the plate face is 2mm
- inflation on all these faces of the boxes or between: 10 layers

Ansys meshing stays stuck. Is there a way to make this easier? I expect the mesh size to be somewhere between 10 and 50 million but it just stays there for hours.

I also tried applying mesh method 'tetrahedron' patch conforming or patch independent, for the air area. It is not helping...


It is stuck at Meshing faces [16%].

[Gweher]

Hello Svetlana,



I didn't manage to download your wbpz from the download link provided (http://sveta.uk.to:2018/~svetlana/CFD/test.wbpz), could you please upload it again ?



10-50M cells can take quite some time depending on the kind of hardware are you using. Did you try increasing the element sizes just to check if it's not a topology issue.