Currently I am doing heat transfer numerical modelling using Ansys Fluent solver. I defined the case as transient with 45 max. iterations/ time step. During the numerical run, the iterations reached to the end while there is no convergence reported. My questions are:
1. Is it necessary to reach to convergence at each time step.
2. If I can not manage to reach to convergence at each time step, is this cause any effect on the solution accuracy.
3. How to reach to the convergence at each time step.
1- Yes. This does not apply to the first few time steps if the initial solution was not obtained by a steady-state simulation, but rather on a global guess.
2- Yes. it will affect.
3- As a rule of thumb, the time step size should be chosen sufficiently small to reach convergence within 5-10 iterations. So if it takes more iterations to converge, it would be more efficient to reduce the time step size than increase the number of iterations.
Whatever problem you wish to simulate, it is advisable to look for a continuous convergence. As Ahmed said before, this condition does not apply during the acceleration period (if any). A punctual achievement of the convergence will affect the speed of the global convergence achievement. To achieve a faster convergence you need to play with the Fluent coefficients which control the numerics.
Thank you very much for your valuable answers Adrian Lungu and Ahmed G. Rahma. I have one concern regarding below statement:
"As a rule of thumb, the time step size should be chosen sufficiently small to reach convergence within 5-10 iterations."
So, to define the correct time step and iterations number we follow trial and error method, or there is a particular method to initially define the best time step and iterations number for each and every problem.
- Badges
- Science topic