CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Coupled and Segregated solver (http://www.cfd-online.com/Forums/fluent/44032-coupled-segregated-solver.html)

 soe March 7, 2007 10:03

Coupled and Segregated solver

I unnderstand the sequential method of solving the conservation laws of physics is fundamenatally different for the two solvers. Basically I solved a steady flow problem of flow over a 3D wing using initial values from the pressure farfield boundary. My initial solver was the segregated solver using the implicit scheme, halfway though when the force coeeficients failed to converge to a steady solution, I switched to the explicit scheme for the coupled solver. I obtained the force coefficients with appreciable stability. Is there any rule in CFD that I cannot change my solver midway into iterations? - I should think not, yeah? This is what I did and I was hoping if anyone can point out any cause for concern with my strategy for obtaining a converged solution.

 Ed March 7, 2007 11:15

Re: Coupled and Segregated solver

I can't imagine it matters. As far as I know after each time step, you should just have a flow field with all the flow properties defined at discrete points and the solver basically restarts from that flow field. How you previously arrived at that flow field shouldn't make any difference.

 soe March 8, 2007 05:37

Re: Coupled and Segregated solver

My sentiments exactly, as far as numerical methods are concerned, a starting value for iterations is required, who's to say the result from the segregated couple is a bad point to start the iteration for the coupled one. But still, I'd like to hear more views from people who have experience rather than conceptual hypothesis like myself for example. Thanks guys.

 All times are GMT -4. The time now is 00:46.