Community
CFD Forum
Welcome to Autodesk’s CFD Forums. Share your knowledge, ask questions, and explore popular CFD topics.
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Simulated Reynolds numbers too small

5 REPLIES 5
Reply
Message 1 of 6
Alexander_Spranger
366 Views, 5 Replies

Simulated Reynolds numbers too small

Hi,

 

I am new to this program and I have problems simulating a moving bar through a water filled Tube.

 

I wanted to begin with simulating different Reynolds numbers to be sure that the settings I picked are working properly - also I thought that moving the bar would be more complicated than just letting the fluid flow around it. But when I tried to simulate flow with a Reynolds number of 70, I noticed that there is no visible vortex - it seems as if the Reynolds number is about 30. Adjusting the mesh and the color scheme didn't do any better.

 

Tube: 1 m long, 22 cm wide, 31 cm high

bar: 1 cm radius, 15.5 cm high (inserted from above)

 

I chose following settings:

 

boundary and initial conditions:

inlet: velocity, 3.15 mm/s, stationary

outlet: pressure, stationary

 

mesh size: automatic -> 36000 elements

 

solver: stationary or transient (tried both)

time step size: 0.1 - 1

inner iterations: 1 - 10

time steps to run: 250 - 2500

 

Did I do any fatal mistake? I used the help function but didn't find something that helped me solve my problem.

 

I really would appreciate any help. Thank you.

 

5 REPLIES 5
Message 2 of 6

Why do you use motion here instead of just streaming the flow around the static bar?

Motion adds some major disadvantages like bigger runtimes and worse accuracy.

Dipl.-Ing. (FH) Marco Müller
Application Engineer Digital Simulation
Mensch und Maschine Deutschland GmbH
www.mum.de/cfd

Message 3 of 6

Sorry for the misunderstanding, but I did try to simulate the flow around the static bar.

Message 4 of 6

 

Hi Alexander,

 

I may be revealing my lack of experience with CFD, but can you clarify what you mean by a "vortex"? What I think of when I read vortex is a swirling of the fluid downstream of the object. Would this really happen at a Reynolds number of 70? Does swirling occur in a laminar condition? (If memory is correct, the transition for pipe flow is around 2200, or maybe higher.)

 

Second question. Since the analysis at 70 gives results that you expect for 30, what results do you get when you run the analysis at 150? 200?  Since the calculated divided by "real world" is 70/30, I would expect a calculated value of 163.3 to give a "real world" value of 70.

Message 5 of 6

Hi John,

 

at a Reynolds number of 70 you would expect the beginning of the Kármán vortex street, so you should at least see two symmetric arranged vortices behind the tube. A real turbulent movement will show up if you have a Reynolds number of about 100.000, but vortices (should) appear already at about 30-100.

I tried a Reynolds number of 140 and it revealed a perfect Kármán vortex street, just how it should be.

 

Is it eventually only a problem of choosing a proper resolution? I tried to use a much finer mesh (about 3.000.000 elements), but my laptop (I'm still waiting for a new PC at work) couldn't finish the simulation because it hadn't enough storage space left.

Message 6 of 6
OmkarJ
in reply to: Alexander_Spranger

Your mesh is very coarse at 36k, have you done mesh sensitivity study? Unless you do that, you can not be sure of your results. Also make sure you have selected laminar model instead of default k-eps model

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk Design & Make Report