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.

Solver not Converging

Please login with a confirmed email address before reporting spam

I'm building a 2D axisymmetric model of a spherical shell with an air-filled interior to try and simulate the heat transfer to the internal air by modeling the outer shell as a heat source. I'm using "heat transfer in fluids" and "laminar flow" as the physics while trying to run a stead state study. However I'm getting the "nonlinear solver did not converge" error, so I'm not sure if my initial or boundary conditions are wrong. I would like some help with this. Thanks!


2 Replies Last Post Jun 28, 2023, 4:33 p.m. EDT
Jim Freels mechanical side of nuclear engineering, multiphysics analysis, COMSOL specialist

Please login with a confirmed email address before reporting spam

Posted: 1 year ago Jun 28, 2023, 3:48 p.m. EDT

Lorenso, it coud be several things including meshing and method of solution depending on your material properties if nonlinear features are added. Also, confused on a 2D axisymmetric model of a spherical geometry. If willing, please attach a sample .mph file, and we might be able to diagnosis better. One suggestion that always holds true is to gradually increase your "load"; especially if you experience problems in converence as you are. The load in this case sounds like the heat addition. For example, you might try a zero heat load to start to make sure you get a steady state equal to the initial state when you solve. That is always a good test. Then perhaps gradually increase the load, and suggest parametric sweep to do that automatically, until you reach the desired heat load. One thought is that the laminar flow (low flow rate implied) may not be sufficient to remove the heat load.

-------------------
James D. Freels, Ph.D., P.E.
Lorenso, it coud be several things including meshing and method of solution depending on your material properties if nonlinear features are added. Also, confused on a 2D axisymmetric model of a spherical geometry. If willing, please attach a sample .mph file, and we might be able to diagnosis better. One suggestion that always holds true is to gradually increase your "load"; especially if you experience problems in converence as you are. The load in this case sounds like the heat addition. For example, you might try a zero heat load to start to make sure you get a steady state equal to the initial state when you solve. That is always a good test. Then perhaps gradually increase the load, and suggest parametric sweep to do that automatically, until you reach the desired heat load. One thought is that the laminar flow (low flow rate implied) may not be sufficient to remove the heat load.

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 year ago Jun 28, 2023, 4:33 p.m. EDT
Updated: 1 year ago Jun 29, 2023, 8:56 a.m. EDT

Hello Lorenzo,

I second James's suggestion that you post your mph file. With that said, if you have a constant heat flux going into a sphere with no way for that heat to leave the system, there is no steady state solution to that problem, and no stationary solver can, nor should, converge.

Jeff

-------------------
Jeff Hiller
Hello Lorenzo, I second James's suggestion that you post your mph file. With that said, if you have a constant heat flux going into a sphere with no way for that heat to leave the system, there is no steady state solution to that problem, and no stationary solver can, nor should, converge. Jeff

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.