Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Coupling between free and porous media flow-impossible to solve in 3D?
Posted Jun 25, 2010, 12:01 p.m. EDT 1 Reply
Please login with a confirmed email address before reporting spam
Hi every body,
I have a model simple including a free domain( a block) and a porous media(a cylinder). The flow is described by the Navier stock equation in free region and the brinkman equations in the porous media. I tried all solver , only solvers indirect and the preconditioner "complete U"could be run, but it was not converged. So, it is impossible to solve this simple coupling?
I attatch the file .mph below if someone want to see the problem
I have a model simple including a free domain( a block) and a porous media(a cylinder). The flow is described by the Navier stock equation in free region and the brinkman equations in the porous media. I tried all solver , only solvers indirect and the preconditioner "complete U"could be run, but it was not converged. So, it is impossible to solve this simple coupling?
I attatch the file .mph below if someone want to see the problem
Attachments:
1 Reply Last Post Jun 26, 2010, 8:42 p.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi, I tried to run your model, but it said it was in a version compiled before 3.5a, so it cannot be opened in 4.0
You could try upgrading to a more recent COMSOL to see if that helps?
More generally, try running at a lower Reynold's number and see if you can get that to converge. If so, then you could use a continuation technique where you slowly increase the Re to what you need, using the previous answer as initial guess to the next.
Regards, John
You could try upgrading to a more recent COMSOL to see if that helps?
More generally, try running at a lower Reynold's number and see if you can get that to converge. If so, then you could use a continuation technique where you slowly increase the Re to what you need, using the previous answer as initial guess to the next.
Regards, John