Floating dok on waves: free surface divergence

Galchenko · September 11, 2014, 02:13

Hello everybody!

I'm trying to simulate wave forces on a floating dok using cfx. The dok is not moving, the wave is coming from the butt end. Computational domain contains half of the body, mesh is structured, hexaedral. The problem oocurs when free surface crosses the body around the half of its height - in this situation there is a water area between 2 walls of the dok. After a few time of simulation(when the wave does't reach the body) there happens a strage rise of water surface between the walls(see img)
freesurf.jpg

I had no problems with simulatings waves on objects like ships, so I used the same ideas to setup this problem:
BC: wave profile on inlet, opening with entraintment on the opposite side and on the top, symmetry, slip wall on the side, no slip wall on dok and bottom.
Solver setup was done according to free surface cfx tutorials. Though I couldn't use volume fraction coupling - solution crashes at first step.
I tried to do calculations usins serial and parallel versions; simualing full dok, not half; refining mesh around dok and free surface; reducing timestep.

Anyway, I couldn't get normal free surface form.. I would really appreciate some help here.. Why crossing free surface incide the domain causes such divergence? Does anyone have any experience or ideas how to solve this problem?

ghorrocks · September 11, 2014, 05:54

How fine is your mesh?

You should try adjusting the free surface modelling parameters. Things like the volume fraction smoothing options might affect this. But I cannot tell you which parameter to adjust, you are just going to have to try them and see what works.

Galchenko · September 11, 2014, 07:22

My mesh around the free surface is about 10 cells per wave height.

There seems to be not many of those parameters.. I mean there is initial volume fraction smoothing with only 1 option - volume-weighted and volume fraction coupling. Also I can add surface tention, but I have read it makes the problem less stable.. Are there any other parameters to vary, which I may not find? I tried to combine these ones in different ways, but the situation didn't change..

ghorrocks · September 11, 2014, 18:04

The decision as to whether to include surface tension is because it has a significant effect on the outcome, not how stable it is. If it is important then you must include it.

10 cells per wave height may not be enough. Try increasing mesh density.

September 11, 2014, 02:13	Floating dok on waves: free surface divergence	#1
Galchenko New Member Galchenko Olga Join Date: Nov 2012 Posts: 16 Rep Power: 13	Hello everybody! I'm trying to simulate wave forces on a floating dok using cfx. The dok is not moving, the wave is coming from the butt end. Computational domain contains half of the body, mesh is structured, hexaedral. The problem oocurs when free surface crosses the body around the half of its height - in this situation there is a water area between 2 walls of the dok. After a few time of simulation(when the wave does't reach the body) there happens a strage rise of water surface between the walls(see img) freesurf.jpg I had no problems with simulatings waves on objects like ships, so I used the same ideas to setup this problem: BC: wave profile on inlet, opening with entraintment on the opposite side and on the top, symmetry, slip wall on the side, no slip wall on dok and bottom. Solver setup was done according to free surface cfx tutorials. Though I couldn't use volume fraction coupling - solution crashes at first step. I tried to do calculations usins serial and parallel versions; simualing full dok, not half; refining mesh around dok and free surface; reducing timestep. Anyway, I couldn't get normal free surface form.. I would really appreciate some help here.. Why crossing free surface incide the domain causes such divergence? Does anyone have any experience or ideas how to solve this problem?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
how to capture free surface cells?	MichaelVS	FLUENT	1	June 10, 2015 02:18
CFD of water waves phenomena-sawteeth at the free surface + slow execution time	nore5	Main CFD Forum	0	May 10, 2013 12:45
Linear analytical solution oto the 2D free sloshing water surface elevation	bearcat	Main CFD Forum	7	August 5, 2011 20:13
Floating ball on free surface validations	Bak_Flow	FLUENT	0	November 10, 2008 16:30
free surface display	novice	Siemens	5	August 4, 2004 01:07

September 11, 2014, 05:54		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	How fine is your mesh? You should try adjusting the free surface modelling parameters. Things like the volume fraction smoothing options might affect this. But I cannot tell you which parameter to adjust, you are just going to have to try them and see what works.

September 11, 2014, 07:22		#3
Galchenko New Member Galchenko Olga Join Date: Nov 2012 Posts: 16 Rep Power: 13	My mesh around the free surface is about 10 cells per wave height. There seems to be not many of those parameters.. I mean there is initial volume fraction smoothing with only 1 option - volume-weighted and volume fraction coupling. Also I can add surface tention, but I have read it makes the problem less stable.. Are there any other parameters to vary, which I may not find? I tried to combine these ones in different ways, but the situation didn't change..

September 11, 2014, 18:04		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	The decision as to whether to include surface tension is because it has a significant effect on the outcome, not how stable it is. If it is important then you must include it. 10 cells per wave height may not be enough. Try increasing mesh density.