CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   help: wave making resistance on submerged sphere (http://www.cfd-online.com/Forums/openfoam/103048-help-wave-making-resistance-submerged-sphere.html)

 amin66 June 9, 2012 07:25

help: wave making resistance on submerged sphere

Hello all
I'm trying to simulate a 3D free surface flow around a sphere. because its first time that i work on free surface, please help me about that problem!

these are my questions:
1. what's the domain(height,weight,width as a function of sphere diameter) in this case
2. Boundary conditions of domain
3. turbulence properties
4. in which reynolds number we have to use turbulent and which one we have to use laminar?
5. solver should be interFoam?
6. some paper and articles about this problem to validate the result like Cd, wave making resistance and some other data.

i'm in force, please help me as soon as you can my friends.
thanks for your attentions and sorry for my bad English.

 amin66 June 9, 2012 07:26

is there any body here that worked on this case before my friends?

 amin66 June 13, 2012 07:41

 nimasam June 13, 2012 08:45

your question is vague, however let me give you some estimation, it may be incorrect
1) i think the domain atleast should be 2 or 3 times of sphere radius in each direction
2) How is your flow? do you have inlet, outlet? or the sphere is in quiescent liquid (it dpeneds on your simulation)

3,4) i suggest you to find papers in this field, then you will find which turbulence model you should use or when it is turbulence!

5) yes, interFoam is a good choice, it has a color function VOF

6) we are not going to research instead of you friend :D

 vonboett June 14, 2012 08:45

Hi Amin,

interFoam is a good choice, I'd confirm. One advice: take good care of the turbulence model, if you want to avoid a lot of discussions and not being a turbulence expert, include as much physics to that as possible. My suggestion is using LES, because then you account for the influence of the sphere geometry on the turbulence

(Wendling I.: Dynamische Large-Eddy Simulationen turbulenter Strömungen in komplexen Geometrien" PhD Tech University Darmstadt 2007 and Ferziger J. H. "Direct and Large Eddy Simulations of Turbulence" in Numerical Methods in Fluid Mechanics Bd 16)

but check that most parts of the grid should be of hexaedral cells for good accuracy. Use LES together with the Dynamic Mixed Subgrid Scale Model (DMM) for example as developed by the LTT Rostock. You need the DMM model to account for the influence of the free surface on the turbulence, where the so called backscatter-effect plays a key role

(Salvetti et al. "Large-eddy simulation of free-surface decaying turbulence with dynamic subgrid-scale models" In: Phys. Fluids 9(8) 1997)

Look for the LTT Rostoch OpenFOAM extensions, together with their simple grid filter that performs the clipping of negative turbulence viscosities based on Taylor series approximations.

(Kornev et al. "A simple clipping procedure for the dynamic mixed model based on Taylor series approximation" In: Communications in Numerical Methods in Engineering 22 2006)

This setup will make a good shift from laminar to transient to full turbulence and it will take care of your geometry and surface, otherwise you will have to explain very well why your turbulence setup (RANS etc.) fits to your case.

If you use an inlet and you don't use a predefined velocity profile but feed in the fluid as a block with same velocity over height, give the inlet some area without wall friction, otherwise you conflict with typical boundary conditions like noSlip (in the U file) at the slope, followed by some area with wall friction so that the turbulent velocity profile can develop before hitting the sphere.

And another thing: If you have an outlet, use zeroGradient as the boundary condition here in your p_rgh file, if you use fixed value 0 as sometimes recommended your fluid gets reflected at the outlet, funnily only if the outlet is situated in the positive coordinate system quadrant. No explanations for this behavior so far but good to know.

 amin66 June 14, 2012 11:32

thanks Dear nimasam
i using a cubic domain with inlet outlet and side walls
i searched alot but couldn't find any paper that says in free surface problem around a sphere, which reynolds number is critical number. so i asked question 4
thank you again for your reply and help

 amin66 June 14, 2012 11:37

Thanks Dear vonboett
i will read articles that you mentioned and do your advised and reply the results. thanks again for your help

 amin66 June 23, 2012 12:07

Friends. i need an article that present the value of drag or Cd for different value of submergence depth(cylinder or sphere) and reynolds number.
this cylinder or sphere moving the free surface.

thanks

 adambarfi July 25, 2012 13:50

you can refer to the White Fluids Mechanics. in this notebook you will find your answers.
I think it was chapter 07: flow past immersed bodies.

 amin66 July 28, 2012 15:54

Hi Mostafa

I read it before but i couldn't see any thing.

 adambarfi July 29, 2012 02:10

Quote:
 Originally Posted by amin66 (Post 367958) Friends. i need an article that present the value of drag or Cd for different value of submergence depth(cylinder or sphere) and reynolds number. this cylinder or sphere moving the free surface. please help if you know or have these articles. thanks
hi Amin,

the Reynolds number: http://adsabs.harvard.edu/abs/1955JPSJ...10..694K
the Drag Coefficient: http://www.grc.nasa.gov/WWW/k-12/airplane/shaped.html

 All times are GMT -4. The time now is 15:49.