I increased the angle of attack to 4° and the amplitude to 1 degree but with reduced frequency to 0.2. The visualization was poor. I used fluid view. What could be the cause?
There can be multiple issues like:-
1) Mesh setting used in the model.
2) Boundary conditions.
3) Numerical errors like numerical diffusion etc.
Ok. Thanks so very much. It worked perfectly well with SA model but SST result was bad for unsteady case. When used for steady state the result was okay.
But what are you talking about specifically? RANS is only steady by definition. URANS Is unsteady depending on external time-dependent forcing but does not describe a physical dynamics of turbulence.
And you know that URANS is somewhat different from the formulations that describe the unsteady details in turbulence?
You have always a statistical solution.
Noah Oyeniran
I don't know what is your backgroud, have you studied the statistical formulations for turbulence?
The averaging is applied on each term of the NSE therefore either the time derivative vanishes (RANS) or its meaning is only physically relevant for resolving an external time forcing (URANS). No representation of a resolved range of time scales is possible.
When using SST-KW turbulence model, a great attention must be paid to the boundary layer. The normalized wall distance Yplus must be lower than "8". This requirement is an important step to guarantee that the first grid layer lies in the laminar sub-layer domain inside the inner layer of the the boundary layer.
Suggested materials:
(1) Statistical Theory and Modeling for Turbulent Flows, Author(s):P. A. Durbin ,B. A. Pettersson
(2) https://youtu.be/w38S4FVYF9E
It can be mesh quality and boundary conditions, sometimes it is not recommended due to the various parameters that the SST model handles.
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators