CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   STAR-CCM+ (https://www.cfd-online.com/Forums/star-ccm/)
-   -   VOF wave getting damped. (https://www.cfd-online.com/Forums/star-ccm/123372-vof-wave-getting-damped.html)

arun7328 September 11, 2013 07:07

VOF wave getting damped.
 
2 Attachment(s)
Dear All,

I am going to work on wave interaction on ships and the dynamic forces for various cases. Just to get the modelling and meshing right for VOF wave and to what comes out of the software for a user defined wave, I am simulating first order waves in a rectangular domain.

The domain is 75m(x), 10m(y which is the direction of wave height), 1m(z direction). The free surface is initialized so that the domain is equally divided. The coordinate of y axis of free surface is 0.

I have defined a wave of length 5 mts, current/wind velocity 1.5 m/s, amp 0.1 mts. The mesh sizes on the free surface is refined using a block of height 0.3 m. The x and z refinement are wavelength/80 and y axis is amp/20. Time step is wave period (2.4*80). I have run the simulation using 50 and 70 inner iterations but the nature of the pressure plot seems to be that of a damped oscillation. Though the waves are not becoming completely flat , which was an issue i had before, it gets damped a little. The point on water level of the wave is 0,0,0. The distance from the wave to outlet is 50mts.

I have also changed the VOF relaxation from 0.9 to 0.5 but there is no change. Can anyone please guide me regarding what is going wrong or what other parameter i could change. I am running unsteady, inviscid model and 1st order solver.

Thanks
Regards
Arun

saeed.barzegar.v December 18, 2013 09:34

2 Attachment(s)
Dear Arun

I had a same problem. I suppose, you are working on old version of fluent? if it is true, you can follow another simple way (by new version of fluent (in ANSYS 13 or 13))
in ANSYS13-fluent, there are options for modeling wave (Airy, stocks,...) and free surface, but it takes times to realize. however, in this way you can simulate completely your problem.
my firs your problem was like "picture 1111" (likes your problem) and my result after modeling in new version of fluent is like "picture 2222" (that wave does not damp)

Also this link can help you for modeling wave in ansys-fluent: http://www.cfd-online.com/Forums/flu...-modeling.html

I hope my suggestion will be helpful.

Jenn January 18, 2016 05:47

Dear saeed,

Would you please also share your experience with inlet and outlet settings?

Thanks a lot.
Jenn

SeaYiTheng February 13, 2017 07:56

Dear Saeed

Hi I am trying to simulate waves too but it dampens out a lot over distance. I am using ANSYS version 16.2. Are you using UDF to generate waves? Can you share with me your UDF code? yithengs@gmail.com
Your help is very very much appreciated :)

Regards,
Yitheng

wester February 13, 2017 14:05

Dear Arun,

I think your result doesn't looks like a wave damping issue. It looks like that there was not enough simulation time for getting the fully developed wave - the initial wave just starts propagating into the calm water area. It might be better to apply the initialization option to fill the whole domain with the wave.

If you don't want a current-wave interaction, start with the zero current/wind velocity.

Also, please check the wave period and length again. It doesn't match.

Regards,
Don

SeaYiTheng February 13, 2017 20:43

1 Attachment(s)
Dear Wester

Attached is the screenshot of the wave I generated, T=1 sec, A = 0.05m using UDF code. As you can see the damping issue is very serious where the wave just disappeared over the distance. Could you please share with me how could I solve the damping issues in my case?
Appreciate your help :)

Regards,
Yitheng

wester February 17, 2017 09:48

Hello SeaTiTheng,

Sorry for the late reply. I can't make sure what was the main damping source just based on the attached image. Frequently, the reflected wave on the outlet boundary might be one of the reasons. Usually, if possible, adding an extra domain (space) is recommended to resolve this wave reflection issue. Also, gradual changes in meshing size and additional numerical damping (or smoothing) is useful to suppress the unwanted wave reflections.

Cheers,
Dong

SeaYiTheng February 17, 2017 10:54

Hi Wester

Thanks for the reply! But i guess reflection might not be the issue here as what I am showing in the picture is the the first three waves that are generated from the moving wall on the right. The first three waves have not reached the left wall yet for reflection to occur i think. As you can see, the first wave is moving from the right to the left and the wave height reduces a lot compared to the waves on the right. Do you have any idea why this happens?
Do let me know if you need any more information on my setup. Appreciate your help :)

Regards,
Yitheng

wester February 17, 2017 11:38

Hi SeaYiTheng,

What's the physical time of the attached image? You have to wait a sufficient time for the fully developed wave field. Assuming a deep water wave, (actually your case looks like an intermediate depth), phase velocity is 1.56 m/s, and group velocity is 0.78 m/s. So you can figure it out how much time needs for the fully developed wave field.

Also did you check if your wave input is valid? It seems like that the your wave input is not appropriate to obtain a very sinusoidal linear wave - not enough water depth and too steep wave profile.

Thanks,
Dong


All times are GMT -4. The time now is 00:37.