[VascoB99]

Hello everyone!

I'm currently undergoing a master thesis in CFD, studying a Surfboard in a non-wave, free-surface scenario.

I am having issues with Numerical ventilation, with the lower section of the board being entirely covered by a layer of air that wouldn't be there in reality, this is affecting the entire calculation. I have tried some of the advice in https://strathprints.strath.ac.uk/75...ning_hulls.pdf but none have worked so far both with low and high y+ treatments.

The problem seems to arise due to geometry, where a recirculation zone happens below the nose of the board and above the free surface line, which the solver cannot resolve. To worsen matters, I am using a code that does not allow me to ramp up velocity in the domain, meaning I have to start every element at 6m/s, the desired speed for my surfboard.

Does anyone have any experience of how to solve this situation? If so your help would be truly appreciated!

The images attached show the initial conditions, as well as a picture of a time step where the air layer is already forming.

Thank you,
Vasco

[arjun]

Originally Posted by VascoB99

Hello everyone!

I'm currently undergoing a master thesis in CFD, studying a Surfboard in a non-wave, free-surface scenario.

I am having issues with Numerical ventilation, with the lower section of the board being entirely covered by a layer of air that wouldn't be there in reality, this is affecting the entire calculation. I have tried some of the advice in https://strathprints.strath.ac.uk/75...ning_hulls.pdf but none have worked so far both with low and high y+ treatments.

The problem seems to arise due to geometry, where a recirculation zone happens below the nose of the board and above the free surface line, which the solver cannot resolve. To worsen matters, I am using a code that does not allow me to ramp up velocity in the domain, meaning I have to start every element at 6m/s, the desired speed for my surfboard.

Does anyone have any experience of how to solve this situation? If so your help would be truly appreciated!

The images attached show the initial conditions, as well as a picture of a time step where the air layer is already forming.

Thank you,
Vasco

Start with a very very small time steps and slowly ramp them. The main issue is that in the start the pressure field is not correct and thus forces are wrong. You might end up in unphysical situations. Increase slowly the time step when the pressure field start to make sense.

[VascoB99]

Originally Posted by arjun

Start with a very very small time steps and slowly ramp them. The main issue is that in the start the pressure field is not correct and thus forces are wrong. You might end up in unphysical situations. Increase slowly the time step when the pressure field start to make sense.

Thank you, I am going to try this, I think there is a big issue with the recirculation zone caused underneath the board nose.

So the recommendation is for me to start with a small velocity and time-step and slowly ramp them up, meeting the CFL = 1 condition. I am also using second-order time discretization, I think this is the way to go to help with this problem.

My time step is currently on the order of 1e-4 s

Edit: I am also using wall functions, but my code supports a Contact line correction option.

[calim_cfd]

Hi,
also check domain resolution.. your mesh seems a bit coarse for your board.. sometimes you don't need to resolve some features of the flow but you might be forced into it not to bump into nonphysical results, convergence issues and ultimately the crash of the model.