Register Blogs Members List Search Today's Posts Mark Forums Read

June 8, 2023, 00:14
#1
New Member

jinhoon park
Join Date: Oct 2022
Posts: 21
Rep Power: 2
hello!
I'm new to openFoam!

Currently, I want to solve the turbulence problem by making a simple model in snappyhexMesh, but I can't converge!

So, I have a few questions for you!

In k-epsilon turbulence, is the initial condition and the mesh influencing the convergence correct?

If that is correct, is it correct to obtain k and epsilon according to the formula below?

Also, is the smaller the size of the mesh, the better the convergence?

As shown in the picture below, after obtaining the initial conditions of k and epsilon, I analyzed
It repeats between residual 0.01 and 0.001 and does not converge..!!

Attached Images
 q1q2.png (26.0 KB, 7 views)

June 8, 2023, 09:25
#2
Member

Amirhossein Taran
Join Date: Sep 2016
Location: Dublin, Ireland
Posts: 42
Rep Power: 9
Quote:
 Originally Posted by tusin111 hello! I'm new to openFoam! Currently, I want to solve the turbulence problem by making a simple model in snappyhexMesh, but I can't converge! So, I have a few questions for you! In k-epsilon turbulence, is the initial condition and the mesh influencing the convergence correct? If that is correct, is it correct to obtain k and epsilon according to the formula below? Also, is the smaller the size of the mesh, the better the convergence? As shown in the picture below, after obtaining the initial conditions of k and epsilon, I analyzed It repeats between residual 0.01 and 0.001 and does not converge..!! please give me some advice!!
Hi Jinhoon,

Answering to these general questions are quite impossible even for CFD specialists, So I would like to suggest that try to tell me more about your problem, maybe geometry or sth like that.

June 8, 2023, 22:23
#3
New Member

jinhoon park
Join Date: Oct 2022
Posts: 21
Rep Power: 2
Quote:
 Originally Posted by amirhosseintaran Hi Jinhoon, Answering to these general questions are quite impossible even for CFD specialists, So I would like to suggest that try to tell me more about your problem, maybe geometry or sth like that.

I would like to analyze the shape attached below by removing air through suction!

Currently u,v,p ,epsilon ,k are
It is the result of convergence down to 1.e-4, but the result is a little different from what I thought, so it seems wrong!

In particular, in the suction part, the fluid is strongly sucked in with y+, but looking at the movement of the fluid, there is a part that goes to y-, so it is questionable!

Can you tell me what is the problem??

I will attach my initial condition, fvsolution and fv scheme!!

Attached Images
 qq11qq22.png (19.1 KB, 8 views) result.png (91.6 KB, 10 views) res.png (18.5 KB, 8 views) u00.jpg (41.3 KB, 6 views) p00.jpg (40.4 KB, 6 views)

June 8, 2023, 22:29
#4
New Member

jinhoon park
Join Date: Oct 2022
Posts: 21
Rep Power: 2
Quote:
 Originally Posted by amirhosseintaran Hi Jinhoon, Answering to these general questions are quite impossible even for CFD specialists, So I would like to suggest that try to tell me more about your problem, maybe geometry or sth like that.

Attached Images
 k.jpg (38.6 KB, 5 views) epsilon.jpg (43.3 KB, 4 views) scheme.png (22.2 KB, 4 views) solution.jpg (36.3 KB, 3 views)

 June 9, 2023, 03:48 #5 Senior Member   Peter Hess Join Date: Apr 2011 Location: Austria Posts: 247 Rep Power: 16 Your outlet channel is very short and you will most likely have a back flow. The velocity at the outlet defined to be zeroGradient. Generally it is a good idea to have inletOutlet BC for the outlet. See chapter 1.51 for more details. https://www.nextfoam.co.kr/proc/Down...tions_of41.pdf Also you can try increase the length of the channel to give the flow the chance to overcome the backflow. You also could need to define the backflow velocity in inletOutlet. And define also the turbulence properties of the backflow could help overcoming the problem.