CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   grid spacing! (http://www.cfd-online.com/Forums/main/6620-grid-spacing.html)

 Bhushan Yadav September 26, 2003 09:27

grid spacing!

Hi my question about grid spacing is still unanswered. Is there anyone who can give some tips over this topic?

 pratap September 26, 2003 10:09

Re: grid spacing!

what is the question ?

 Bhushan Yadav September 28, 2003 08:29

Re: grid spacing!

Hi Pratap Thanks My question is for laminar viscous flow over a flat plate or for that matter on any other 2D configuration what is the criteria to choose first grid spacing, stretching function etc.? Does it differ for subsonic flow and hypersonic flow?

Regards Bhushan

 Sun October 1, 2003 02:49

Remarks on grid spacing

hello, the grid size strongly depends on your problem and which region you are interest.

For the flat wall, the minimum height of grid size is about 0.2/sqrt(Re) to get a good solution at the distance of x=1 from the leading tip of the flat plat. For the flat wall, a high aspect ratio is helpful to stablize the solution, but not essential.

For more general problems, you may put 5 grid layers within the boundary layer.

Sun

 derrek October 3, 2003 10:51

Re: Remarks on grid spacing

Hello Sun:

I am just curious how you determine 0.2/sqrtRe at x=1? I am not disagreeing ust curious?

cheers,

derrek

 Sun October 4, 2003 08:09

Re: Remarks on grid spacing

The thickness of the boundary layer over a flat wall is about 1/sqrt(Re). If you add 5 grid layers there, then the grid size is 0.2/sqrt(Re). Centainly, you may choose finer grid size.

Once I did numerical tests for the problem a few years ago. I noticed that the grid size should be about the value or finer to be able to get a good friction force. The finer grid size improve the accuracy very little.