I'm conducting simulaion for enclosed geometry (triangular) natural convection. It is for low Gr numbers, of Gr~1e5 (at least I call it low). I'm recreating somebody elses results and have had success with the solving using segeregated solver with the solution converging with a high number of iterations of a FEW thousand.

However when I use a coupled solver it diverges badly within the first hundered iteration even with a very small Courant number of 0.1, then slowly it converges over the next few thousand iterations, but nowhere near the speed with which the segerated solver converges.

This is confsing as the manual says that a coupled solver converges quicker.

Other points: using the exact same model composed of 7330 quad triangular elements.

thanks in advance

Allan

Hi,Everybody: I want to model air, propane vapor and propane liquid mixing. This problem includes three material so I want using the Euler model .But in Fluent Euler model regard the second phase (expct for the first phase ) as droplet ,but my second phase (propane vapor)is continous phase. I must add the species equation to fluent using udf. Now I can't find a Macros to connect different matrial .

Anyone have idea or experience in writing udf about speiecs equation . Help me

thanks in advance sunny

Dear sunny

I am also facing the same problem. But as per my knowledge, you cannot integrate the species transport and Eulerian model in Fluent. If you find any useful solution to this problem may be through udf, you please forward it to me.

Regards Karthick

Dear Allan

Coupled solver will take more time than segragated solver for each individual iteration becuase it has to solve all equations simultaneously unlike segragated solver where equations are solved sequentially. Moreover its fine that you are getting convergence. normally people face convergence problem with the coupled solver in the wanting situations!!

thanks and regards Prasanth