Fluids

Fluids

Courant number

    • A1234
      Subscriber
      I simulate vof model. My first time step courant number is 0.1 but second is 2.24 and i have courant number upper 1 till 30 time step then decrease. Is my solve wrong and need to get smaller time step?
    • RK
      Ansys Employee

      Hello, 


      What is the end goal of your simulation?


      Regards,


      Rahul

    • Karthik R
      Administrator
      Hello Courant number is generally used to ensure a stable solution. Based on the velocity scales and your grid size, picking a certain courant number will ensure a certain time step step size. The smaller the courant number, the more stable your solution would be, and smaller your simulation time step. Generally rule of thumb is to pick time steps such that the courant number is closer to 1. In order to understand whether your solution is correct, please check the physical time scales of your problem. Your time step choice should be lower than this physical time step. This will ensure your that your simulation results are correct.
      Hope this help.

      Best Karthik
    • A1234
      Subscriber

      thanks for answering


      but what the mean of "Your time step choice should be lower than this physical time step."?


      what is physical time step?


      2.How to increase the time step without increasing courant number? at fixed mesh size 


      That is, which settings in the fluent affect the courant number

    • Karthik R
      Administrator

      The physical time scale is based on the physics of your problem. For example, if you are looking at transient condution, depending on the length scale and the thermal diffusivity of the material, there is a physical time scale responsible for conduction. If you are taking time-steps larger than this conduction time scale, it is quite likely that you are not capturing the physics correctly.


      To answer your second question - for a fixed mesh and velocity scale in your problem, you cannot increase your time-step without adversely affecting your Courant number (at least). It is very much dependent on the min grid spacing and velocity scale of your problem.


      Thanks.


      Karthik

Viewing 4 reply threads
  • You must be logged in to reply to this topic.