Taking too much computational time to converge #28259
Replies: 2 comments 6 replies
-
|
Hello It would go a lot faster if you did not fail so many time steps.
to try to find out why it's failing every other time step. Guillaume |
Beta Was this translation helpful? Give feedback.
-
|
I did the |
Beta Was this translation helpful? Give feedback.
-
|
looks like one of these options cannot be used with your current solver options |
Beta Was this translation helpful? Give feedback.
-
|
@Emmanuel-324 I don't know how big is your domain and how much strain you are applying, but from my experience that behavior is due to the time step being too large. Try reducing the time step and see if it can avoid the unconverged steps. For the crystal plasticity model, my experience is that aiming for 0.01% strain increase, or less, for each time step will lead to good convergence. It depends on the constitutive model though. As @GiudGiud says, regular unconverged steps will lead to a very slow simulation overall. |
Beta Was this translation helpful? Give feedback.
-
|
My domain is a 2D 300 by 300 mesh. I am applying a constant displacement. I am hoping to get a strain of at least 0.008. Thanks, I will try your suggestion. |
Beta Was this translation helpful? Give feedback.
-
|
Did you try taking smaller time steps? |
Beta Was this translation helpful? Give feedback.
-
|
When you say smaller timesteps do you mean smaller timestep interval or the total time of the simulation. Below is what my executioner block looks like |
Beta Was this translation helpful? Give feedback.
-
|
We mean intervals. |
Beta Was this translation helpful? Give feedback.
-
I have a bicrystal script that takes too long to converge to a meaningful output. Do you have any suggestions to help resolve that?
Below is the script
Below is a snippet of the output
Beta Was this translation helpful? Give feedback.
All reactions