[Rebecca]

I'm pretty new at GridGen. I'm trying to setup a particular analysis area inside a larger airflow. The analysis area is constructed as an extruded structured block (cylindrical). The larger area is an unstructured area. I continue to have problems and am trying a variety of solver configurations to make it a decent mesh for Fluent. Has anyone done this before, and would you provide some tips for me?
Thank you,
Rebecca

[rmatus]

Hi Rebecca:

What sorts of problems are you having? Not being able to extrude a mesh? Problems in the transition from the structured grid to the unstructured grid?

If you can give some specifics, I will see if I can help.

Rick

[Rebecca]

I created the structured block by extruding a circle with 0.159 m dia to 0.5 m in 1 cm steps. It turns out okay with no errors.
I can create the unstructured block (a huge wind tunnel) with 30 aniso layers, 10 modtopo layers, 0.9 boundary decay, Equi-Volume 0.5, Equi-angle 0.5, centroid 0.5, max angle 150. (The turret inside the tunnel has structured domains.)
The problem is creating the unstructured block with the structured block inside. I was very careful on this repitition of unstructured block solving, but I still have the structured block surrounded by bad cells - "get_blk_vol_unscell_iblank: bad cell." So, I think it's not matching up with the domain created by extrusion very well.

[Rebecca]

Rick,
I also tried the same thing, but making the domains of the area to be extruded and the surrounding areas unstructured rather than structured, and the final product after the block creation for both the extruded area and the overall fluid area still has bad cells surrounding the extruded block.

[Rebecca]

Rick,
I forgot to say thanks for looking into it. I get distracted with all these things I'm trying to learn for the first time. Thanks.
Rebecca

[jorkolino]

Hi, did you had advance in your problem with bad cell? How can you identify the bad cells on the screen? I don't see a way Gridgen will allow you to do that. I just need to find out where they are, instead the console window prints bad cell IDs.

I checked now and this issue appears only at pyramidal interface, and all pyramidal interfaces, exactly. They are good looking, yet GG reports get_blk_vol_unscell_iblank: bad cell (id....)

I wouldnt care about them but my software seems to have difficulty converging. The cell quality is good, when reported by GG. Skewness is well within limits, everything. It is just the console messages when examining the grid. Strange. bad cell, bad cell, bad cell, bad cell....