URL has been copied successfully!

(1) Divergence in OpenFoam simulation (2) Speeding up simulation

  • (1) Divergence in OpenFoam simulation (2) Speeding up simulation

    Posted by Grace Aquah on January 9, 2024 at 4:49 pm

    (1) Please, does anyone know the reason why a simulation will diverge in openFoam when the velocity is reduced, for instance, from 1 m/s to 0.1 m/s? Domain size is 300 microns by 600 microns by 300 microns, and dx is 5 microns.

    (2) Also, does anyone know how to speed up a simulation in openFoam that is running very slowly?

    Both simulation cases are two-phase flow cases using interFoam solver.

    Sérgio Cavaleiro Costa replied 6 months, 4 weeks ago 2 Members · 3 Replies
  • 3 Replies
  • Sérgio Cavaleiro Costa

    Member
    January 9, 2024 at 9:15 pm

    Hello Grace! I am not aware of what you are trying to solve, but this is a very small domain where you find a velocity considerably high (for the domaine size). Look at how much time a particle need to take from the entry to the outlet of your domain. It is a very small time interval. Furthermore, this domain is discretized, so the cell elements may be very close to the minimum size of the assumption of a continuum.

    Relatively to the convergence, you should look to your Courant number (you may also find the name CFL number). This may give you a clew about the reasons to the slow convergence, but I also suspect it is related with my previous points.

  • Grace Aquah

    Member
    January 9, 2024 at 10:52 pm

    Thank you for your response. Please what do you mean by “the cell elements may be very close to the minimum size of the assumption of a continuum.”?

  • Sérgio Cavaleiro Costa

    Member
    January 10, 2024 at 9:01 pm

    The Navier-Stockes equations derivation assume that they are supported in a continuum domain. You can know more about that size based on the Knudsen number (https://en.m.wikipedia.org/wiki/Knudsen_number#:~:text=The%20Knudsen%20number%20(Kn)%20is,a%20body%20in%20a%20fluid.)

Log in to reply.