Learning Forum

[sergiofr]

Hi,

I am simulating the water wave propagation and interaction with a subsea structure. For some reason I do not know the simulation runs for several time steps without problems and in some point the VOF CFL starts rising up to values of 5.0 when the simulation blows up.

What is the reason that makes the VOF CFL value rise up? How is it CFL VOF value calculated? I have tried looking into the documentation but could not find it.

Best Regards,

Sergio

[Surya Deb]

Hello,
You might want to check this link for details [https://ansyshelp.ansys.com/account/secured?returnurl=/Views/Secured/corp/v211/en/flu_ug/flu_ug_sec_vof_using_td.html?q=vof%20sub%20time%20stepping]
Also, you can test Dynamic Adaption based on VOF gradients to capture the gradients across the interface properly. Check this link [https://ansyshelp.ansys.com/account/secured?returnurl=/Views/Secured/corp/v211/en/flu_ug/flu_ug_adapt_refine-coarsen.html?q=Dynamic%20Adaption]
Regards SD

[sergiofr]

Thanks for your suggestions and sorry for my late response but I have been working on it. I cannot perform Dynamic Adaption because my mesh is large, so the computational time would be very large.
Instead, I have been doing some findings in the simulation. Before the VOF CFL rise up to values higher than the specified one (0.15), it is possible to see how the GLOBAL CFL rise up to values which are higher than the specified one (0.3). Figure below show how GLOBAL CFL evolves along the X-direction (Wave propagation direction). As you can see the CFL value near the platform is higher than 0.3 which is the specified maximum. Indeed, when the simulation continues this value keeps rising up to 250 when the simulation blows up.
The values shown in the graph above are not in accordance with those shown by Fluent in the console output when the simulation runs
Also, I would like to point that the simulations converges to the specified criteria even when whte CFL values are higher than expected in 10 to 15 iterations. So, I would say it is not a problem with the mesh that has a good quality (Orthogonal quality values -> Min: 0.2 - Max: 1.0- Mean: around 0.8)
I do not know why it is happening. According to documentation you provided to me and to the section "36.14.4. Multiphase-Specific Time Stepping", the time step used should be the minimum of all the time steps calculated for different physics in the mesh. So, there should not be any cell where the CFL value where higher than 0.3.
Here is a quick summary of my simulation setup:
Solver type: Pressure based
Tim: Transient
Gravity: On
Muliphase: VOF - Open Channel Flow - Open Channel Wave BC - Inerface: sharp - Interfacial Anti-diffusion: enabled - Formulation: explicit
Viscous: Laminar
PV Coupling sheme: PISO
Gradient: LSCB
Pressure: Presto!
Momentum: Second Order Upwind
Volume Fraction: Compressive (I have also tested with Geo-Reconstruct)
Transient Formulation: First Order Implicit
Warped-Face Gradient correction: enabled
Hybrid Rhie-Chow term: enabled

Here, I attatch to you an snapshot of my mesh

Why the prediction algorithm is not working properly? Is there something wrong in my simulation setup?

Best Regards Sergio