Turbulent shallow water flow

phy4me · November 23, 2017, 04:35

Hello everyone,

With an algebraic (Mixing lenght) turbulent shallow water flow, I can not see recirculation developed around an obstacle. My simple question is: can the number of mesh have an effect on the capturing of turbulent behavior? or I have to use other models?

Thanks.

FMDenaro · November 23, 2017, 07:51

Please post the plot of the solution

phy4me · November 23, 2017, 09:54

with pleasure

FMDenaro · November 23, 2017, 11:16

Well, the grid resolution appears very coarse but that depends on the Reynolds number of the flow. Are you sure that the run is convergent?

However, if you are using RANS then you would not be so surprised that you do not capture a great recirculating region behind the bluff body.

phy4me · November 23, 2017, 11:36

I'm not sure for the convergence, because I'm not the responsable of the runs. I'll check this point.

Indeed, I use the RANS model. Is this the problem ?

Thank You

FMDenaro · November 23, 2017, 11:46

Quote:

Originally Posted by phy4me

I'm not sure for the convergence, because I'm not the responsable of the runs. I'll check this point.

Indeed, I use the RANS model. Is this the problem ?

Thank You

What is the goal of the simulation? If you want to observe the vortex shedding, RANS is not the correct formulation.

phy4me · November 23, 2017, 11:52

We want to develop a numerical scheme based finite volume method for the tubulence shallow water flows. We start with the RANS model involving many turbulence models (Minxing lenght) and K-epsilon model. As an example we try to compare all on a wake turbulent.

nourlil · November 27, 2017, 11:20

Hello,

Firstly i think mesh size around obstacle don't have good quality, you can keep actually size of mesh on the geometry but you have to refine it around obstacle.

mesh.PNG

result.PNG

piu58 · November 27, 2017, 14:32

I don't recommend triangular meshes because they are very diffusive. Your problem may be meshed without any problems with hex elements.

nourlil · November 28, 2017, 04:01

Hi,

Due to simple probleme and geometry he can do it with 3D so i recommand you to use tetraedral element and insert prism layer on the wall and around obstacle to obtain the effect of viscous forces near the wall.

phy4me · November 28, 2017, 04:11

Thanks all for the replys

Nourlil, I appreciate your remark to refine near obstacl, I well do it and post results.

Piu58, I have all time the problem of numerical dissipation but I never had the idea that the rectangular mesh is little diffusive. Sadly, I work with a personal code, is mush diffuclte for me tu converte to a rectangular mesh only for this problem.

nourlil · November 29, 2017, 10:51

Quote:

Originally Posted by phy4me

Thanks all for the replys

Nourlil, I appreciate your remark to refine near obstacl, I well do it and post results.

Piu58, I have all time the problem of numerical dissipation but I never had the idea that the rectangular mesh is little diffusive. Sadly, I work with a personal code, is mush diffuclte for me tu converte to a rectangular mesh only for this problem.

No problem you are welcome, which soft do you use to simulate ? if you want i can help you to simulate your calculation with static and dynamic obstacle.

Best,

November 23, 2017, 04:35	Turbulent shallow water flow	#1
phy4me Member C. Ziyad Join Date: Oct 2010 Posts: 37 Rep Power: 15	Hello everyone, With an algebraic (Mixing lenght) turbulent shallow water flow, I can not see recirculation developed around an obstacle. My simple question is: can the number of mesh have an effect on the capturing of turbulent behavior? or I have to use other models? Thanks.

November 27, 2017, 14:32		#9
piu58 Senior Member Uwe Pilz Join Date: Feb 2017 Location: Leipzig, Germany Posts: 744 Rep Power: 15	I don't recommend triangular meshes because they are very diffusive. Your problem may be meshed without any problems with hex elements. __________________ Uwe Pilz -- Die der Hauptbewegung überlagerte Schwankungsbewegung ist in ihren Einzelheiten so hoffnungslos kompliziert, daß ihre theoretische Berechnung aussichtslos erscheint. (Hermann Schlichting, 1950)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Issues on the simulation of high-speed compressible flow within turbomachinery	dowlee	OpenFOAM Running, Solving & CFD	11	August 6, 2021 06:40
Problem with divergence	TDK	FLUENT	13	December 14, 2018 06:00
setting for shallow water	nurul aqilah	Fidelity CFD	0	December 7, 2015 20:44
Is it a turbulent flow or laminar flow?	ringtail	Main CFD Forum	9	January 22, 2015 19:52
Terrible Mistake In Fluid Dynamics History	Abhi	Main CFD Forum	12	July 8, 2002 09:11

November 23, 2017, 07:51		#2
FMDenaro Senior Member Filippo Maria Denaro Join Date: Jul 2010 Posts: 6,773 Rep Power: 71	Please post the plot of the solution

November 23, 2017, 11:16		#4
FMDenaro Senior Member Filippo Maria Denaro Join Date: Jul 2010 Posts: 6,773 Rep Power: 71	Well, the grid resolution appears very coarse but that depends on the Reynolds number of the flow. Are you sure that the run is convergent? However, if you are using RANS then you would not be so surprised that you do not capture a great recirculating region behind the bluff body.

November 23, 2017, 11:36		#5
phy4me Member C. Ziyad Join Date: Oct 2010 Posts: 37 Rep Power: 15	I'm not sure for the convergence, because I'm not the responsable of the runs. I'll check this point. Indeed, I use the RANS model. Is this the problem ? Thank You

November 23, 2017, 11:52		#7
phy4me Member C. Ziyad Join Date: Oct 2010 Posts: 37 Rep Power: 15	We want to develop a numerical scheme based finite volume method for the tubulence shallow water flows. We start with the RANS model involving many turbulence models (Minxing lenght) and K-epsilon model. As an example we try to compare all on a wake turbulent.

November 27, 2017, 11:20		#8
nourlil New Member France - Alsace Join Date: Oct 2017 Posts: 13 Rep Power: 8	Hello, Firstly i think mesh size around obstacle don't have good quality, you can keep actually size of mesh on the geometry but you have to refine it around obstacle. mesh.PNG result.PNG

November 28, 2017, 04:01		#10
nourlil New Member France - Alsace Join Date: Oct 2017 Posts: 13 Rep Power: 8	Hi, Due to simple probleme and geometry he can do it with 3D so i recommand you to use tetraedral element and insert prism layer on the wall and around obstacle to obtain the effect of viscous forces near the wall.

November 28, 2017, 04:11		#11
phy4me Member C. Ziyad Join Date: Oct 2010 Posts: 37 Rep Power: 15	Thanks all for the replys Nourlil, I appreciate your remark to refine near obstacl, I well do it and post results. Piu58, I have all time the problem of numerical dissipation but I never had the idea that the rectangular mesh is little diffusive. Sadly, I work with a personal code, is mush diffuclte for me tu converte to a rectangular mesh only for this problem.