CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   mapping with Thompson scheme (http://www.cfd-online.com/Forums/main/515-mapping-thompson-scheme.html)

 Juliet February 4, 1999 04:30

mapping with Thompson scheme

Hi,

I am a beginner of CFD. I tried to use thospson scheme to map uniformly spaced circles in physical space into a uniform rectangular grid in the computational plane. I specified boundary conditions with polor coordinate r=0~1, then transfer into cartesian coordinate x and y. But if the domain is 0~pi/2, the results of x and y are correct, uniformly spaced circles can be achieved. If the domain is 0~2pi, the results will be changed, the space of radial direction is not expected. Would anybody give any suggestion about it? Thanks in advance!

Juliet

 Eiyad February 6, 1999 00:58

Re: mapping with Thompson scheme

Hi Juliet, There is an easier way than the Thompson scheme. This way is called mathematical grid generation, where you map your grid using some described mathematical function.

As for your problem, you may didn't use the appropriate periodic boundary condition.

If you have any question just feel free to ask me through thisa email

 Eiyad February 6, 1999 00:58

Re: mapping with Thompson scheme

Hi Juliet, There is an easier way than the Thompson scheme. This way is called mathematical grid generation, where you map your grid using some described mathematical function.

As for your problem, you may didn't use the appropriate periodic boundary condition.

If you have any question just feel free to ask me through thisa email