[michaels]

Hi everyone,

a question concerning wave generation:
how many cells would you recommend over the waveheight/wavelength concerning the calculation mesh?
Is there another way of plotting the free surface elevation except using the xy-plot option and an iso surface as the part to be shown?

Regards

Michael

[Vinicius]

Hi Michael,

20 cells per wavelenght is a good number, and you can use this for the height too.

The isosurface polt is a good way to see the water elevation, but yuo can also cut a plane in your domais and plot the mass fraction of water and see it laterally.

regards,

[michaels]

what exactly do you mean by "cutting a plane"? I´m only using starccm+ for 3 weeks and i´m not an expert yet

I´m still having the problem that the waves don´ t keep their initial amplitude, they change their height...

[Vinicius]

It depends what boundary conditions you are using and what order of wave.

How did you set up your model?

[michaels]

waves are first order; amplitude 0.0565m and period is 1.05secs.
inlet boundary is velocity inlet, sides are velocity inlet, outlet is pressure outlet and top and bottom also velocity inlets.
which conditions would you use? mine are the same as in the tutorial that´s why i don´t know what the problem might be
maybe try fifth order waves?

[Vinicius]

You boundary conditions are correct.

Be sure that you set the velocities at inlet as components in the Velocity specification (Physics Conditions > Velocity Specification node, and change the Method property to Components in the Properties window) and in the Physics values node, change to field function Velocity of "Your Wave".

The first order waves are good when you have the relation between wave height/wavelenght < than 0.1

Maybe you can try the fifth order waves.

regards

[michaels]

all velocity inlets are specified as components, also the initial conditions are set correctly for "my wave".
pressure is also splitted in volume fraction and hydrostatic pressure of "my wave" as it is done in the tutorial...
i´ll give the fifth order waves a try at the weekend and let you know about the results
are there any things i should pay attention to concerning the mesh and the growth of the cell sizes? i already have a fine mesh near the free surface...
thanks so far for your help

regards

[Vinicius]

About the mesh, use trimmed cells and refine in the free surface region, as you already did.
As you are just simulating the waves, you don´t need the pay attention with interaction between a body and the fluid, but when ou were working with a boat, for example, you must refine the body and have a good control of the mesh growth around it and a wake refinement.

[michaels]

okay, i´ll see what i can achieve over the weekend and then post pictures of my mesh and the surface elevation plots i´ll also check fifth order waves...
thanks again so far...

[michaels]

Hi Vinicius

spending some time on mesh generation and mesh quality i got the waves to work perfectly....5th order and also 1st order waves...
thanks a lot anyway

[arun7328]

Dear Micheal,

I know it is too late but can you please tell me the mesh refinement you used for generating your wave. I am trying to simulate a first order wave with amplitude of 0.1m and length 5 mt. The total domain is 75X10X1. y axis forms the direction of amplitude. I have used a block of height 0.3 mts to refine the free surface. The mesh sizes in x and z direction is 0.0625(5/80) and y axis is 0.01 (0.2/20). I am getting a good comparison of pressure and velocity. My concern is the total cell count is very high. I am a wary because of this. When I will be working with ship wave interactions the total mesh can be too high increasing the time. Also if the wave is so sensitive to mesh size, even for a small of amplitude or wavelength would require remeshing.
Could you please share your knowledge on wave simulation please?

Thanks
Regards
Arun
PhD student
FSI group
Southampton University