[CIG-SHORT] [EXTERNAL] How to set bias fine size {a} coarse size {b} ?

[Brad Aagaard]

Bei Xu,

The gradient.jou script runs for me using CUBIT 15.3. I did notice that 
you only apply the bias to the extension of the fault surface. When I 
meshed the surrounding surfaces I end up with some very coarse surface 
cells. I don't think this is what you want; I think you want to apply 
the bias to the other surfaces that have bias on their edges to get a 
nice smooth gradation in the cell size.

If the mesher is crashing on you, then I would use the "top" command (or 
another process monitor) to watch how much memory the mesher is using. 
You could be running out of memory.

We strongly recommend debugging your workflow using a coarse mesh and 
then working your way to the final fine mesh for production runs. In 
this case, I would debug the mesh sizing issues with a coarser cell size 
(something like 400-500m) on the fault. Once you are happy with that 
mesh, then use finer sizes. You should only have to change the fault 
size parameter.

Regards,
Brad


On 04/27/2018 01:14 AM, xu bei wrote:
> Hi, everyone
> 
> I can construct a geometry fault model with Trelis. But I failed to mesh 
> the fault surface into small grid (cell size is 100 m on the fault 
> surface). I wrote a gradient.jou file, but the Trelis crashed when 
> running this script. I think there are some problems when I set  bias 
> fine size {a} coarse size {b}. I attached the .jou files. Hope someone 
> would give me a hand to help me solve the problems.
> 
> Best regards,
> 
> Bei Xu
> 
>