Fluids

Fluids

Topics related to Fluent, CFX, Turbogrid and more.

Definition of viscous force in UDF

    • Bastyx
      Subscriber

      Hello everyone,

      I am writing my master thesis and I am trying to simulate 1DOF oscilations of different cross-sections. To do so, I am using 6DOF UDF (as can be seen below in attached img), but the definition of viscous force (F_STORAGE_R_N3V(f, t, SV_WALL_SHEAR)) gives me SIGSEGV error everytime I try to start the calculation. If I commented the viscous force out (as seen in attached image of udf) it works well.

      My supervisor says this definition of viscous force worked for him last year for different project, but doesnt work now. I am using Fluent 2021 R2, but i dont think its version related problem.

      Does anyone have similar issues or can somebody help me how to get the viscous force work?

      Thanks in advance and if anyone has any questions or more details is needed, feel free to ask.


    • Rob
      Forum Moderator
      What happens if you run a time step then turn it on?
    • Bastyx
      Subscriber
      Hi Rob, thanks for the answer.
      I should specify that I am using dynamic mesh method and only layering method. I have the mesh and all set up and it works fine without the viscous force in the UDF.
      To answer your question. Before I use dynamic mesh (with the UDF) I have computed few thousands of timesteps without the movement of the body to develop the pressure and velocity field. Then I "free" the body with the compiled 6DOF UDF. If the UDF contains the viscous force, it crashes with the SIGSEGV errror.
    • Rob
      Forum Moderator
      Not all storage values are computed with the initial condition. I was wondering if that needs a timestep to be completed to fill that variable.
    • Bastyx
      Subscriber
      Okey, I see. But apparently its not the case with F_STORAGE_R_N3V. Can you think of anything else what could cause the error?
    • Rob
      Forum Moderator
      SIGSEV often means the data isn't there, ie it's not defined. What models are you using that your supervisor didn't? Or vice-versa.
    • Bastyx
      Subscriber
      We both used SST k-omega model.
    • Rob
      Forum Moderator
      Both single phase?
    • Bastyx
      Subscriber
      Yes, both single phase.
    • Rob
      Forum Moderator
      I wonder if the macro changed, it's not documented and some of the storage terms are altered between versions. Ask your supervisor to find the header file it came from in the working version of Fluent, then look in the same file in your version: you may find there's an extra term needed or the like.
Viewing 9 reply threads
  • The topic ‘Definition of viscous force in UDF’ is closed to new replies.