CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   multigrid blues (https://www.cfd-online.com/Forums/main/8930-multigrid-blues.html)

parth March 29, 2005 04:45

multigrid blues
 
Hello all, I am trying to write a code for FullMultigrid , I have a small doubt, When performing relaxation on coarser level , the equations are:

A_p x Phi_p + Sum( A_l x Phi_l) = Src_p + AdditionalSource;

AdditionalSource = (A_p x Phi_p + Sum( A_l x Phi_l) -Src_p)_by_restrictedvalues -restricted_Residual,

This remains fixed during this sweep.

Then I transfer my values to further coarser level.

Now my doubt is, what shall I do , when I transfer the correction back to finer level from coarser level. I mean the corrections are transferred and I have now the new values of the variables.And I shall do _post_relaxation_sweeps_ here, but what happens to this restricted_Residual, required in the additional source.

I am confused about this value. How do we construct the equation now for _post_relaxation_sweeps_????

Do I use the last restricted_Residual value (calculated when transferring from fine to coarse),

And what about the fluxes on the faces, how do we get them when we want to do post_relaxation_sweeps_ , do I calculate them from these corrected values?


versi April 1, 2005 05:33

Re: multigrid blues
 
You can keep the last restricted_Residual value (calculated when transferring from fine to coarse) during the upward sweep (the upward sweep refers the one from coarse to fine) But the fluxes need to be calculated again in the up sweep.

parth April 1, 2005 06:24

Re: multigrid blues
 
thanks for your reply, i really appreciate,

so when the variables(u v w p) are corrected the fluxes are calculated again based on corrected values,

and when the we go fine to coarse, the fluxes the same as of finer level (after restriction) ..correct??


All times are GMT -4. The time now is 08:35.