|
[Sponsors] |
January 3, 2022, 16:02 |
2D or 3D turbulence benchmark models
|
#1 |
New Member
Naive CFD
Join Date: Jun 2021
Posts: 25
Rep Power: 5 |
Hi everryone, I have a question about turbulence. Since I am new to it, I am trying to benchmark a few models in OpenFOAM. But I after researching online. i.e, NASA turbulence repository. Most of the test cases they have shown are in 2D. Since turbulence is a 3D effect, is it enough to validate turbulence models against the 2D test cases of NASA repository i.e., 2D airfoil or backward-facing step.
I am looking in Spalart-Allmaras and K-e in particular. It would be really helpful if you can comment on it. |
|
January 3, 2022, 16:19 |
|
#2 | |
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 6,882
Rep Power: 73 |
Quote:
There are some posts in this section you can look at, similar discussions are highlighted. However yes, turbulence is 3D and unsteady by definition. What you see in the 2D approach is a consequence of a statistical mean performed on the NSE. For statistically steady flows, with a homogeneous direction, you can solve the steady RANS equation in 2D. |
||
January 3, 2022, 16:41 |
|
#3 |
Senior Member
Lucky
Join Date: Apr 2011
Location: Orlando, FL USA
Posts: 5,754
Rep Power: 66 |
Benchmarking and validating are two different things. If you run a backward 2D facing step, you should get the same solution as NASA (your benchmark).
As mentioned already, you doing RANS and solving for statistics. "Turbulence is 3D" is a statement about its real dynamics in spacetime. The probability of getting heads (or tails) from flipping a coin is 50% but that doesn't mean that if you flip a coin you will get half a head. Flipping a coin is a binary outcome, you either get heads or tails. The statistics of a coin do not state that the coin has half a head and half a tail and neither does it contradict the fact that a real coin has two faces (one heads and one tails). |
|
Tags |
turbulence |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Validation of homogeneous turbulence with constant shear for RANS models | GeorgS | OpenFOAM Running, Solving & CFD | 2 | August 30, 2018 16:11 |
Adding turbulence models to (reacting)TwoPhaseEulerFoam | lavdwall | OpenFOAM Running, Solving & CFD | 0 | February 3, 2017 04:04 |
y+ range for turbulence models | babri | FLUENT | 1 | August 30, 2016 18:39 |
in k-epsilon wall function approach high Re turbulence models: question of velocity | romant | OpenFOAM Programming & Development | 6 | May 26, 2016 10:14 |
Advice about turbulence models? | mqsim | OpenFOAM Running, Solving & CFD | 6 | April 27, 2016 08:50 |