SST transition
 

Hi,
I'm rather new to this forum but I wanted to share with you an adaptation I made of the gamma-Re_theta turbulence transition model for OpenFOAM 4.x.
I used the tutorial available here, from the Chalmers Swedish University of Technology, but the (great) work they did was only for OpenFOAM 2.0.x.

I still need to validate the results, if you could give me your feedback that would be awesome!

To use it, the easiest is to have you own version of src/ in $WM_PROJECT_USER_DIR, and so just git clone git@gitlab.com:tilasoldo/openfoam_share.git. Your version of the executables will be used, since $PATH is defined with your src path first.
Then compile the source: go in $WM_PROJECT_USER_DIR/src and run ./Allwmake.

To use it, you must define in addition to what is needed for the k-omega SST:
- initial conditions, fvSchemes and fvSolution for im (transition ratio) and Ret
- constant/transitionProperties, which only contains the freeStreamU value.

I also included in git the test case available in the tutorial, which sums up the usage of this solver.

Let me know if you have any trouble to run or compile it, I could probably help :)

Enjoy and have a nice day,
X. Lamboley

Hi Tilasoldo,

I am also new to these forums and this is actually my first post here. I am currently working on implementing Menters new transition model (One equation LCTM 2015) in OpenFOAM 4.1. so it would be very helpful to have a peek how you managed to deal with OF4.x template metaprogramming.

However, the link you provided isn't working anymore so is it possible to upload the model again? It could be interesting to compare the results of these models aswell.

With regards
Unto

Hi Unzreal,

I just retested my link and it works. However, it is a git link, you can download the source by running git clone git@gitlab.com:tilasoldo/openfoam_share.git in a terminal.
If you prefer, I can provide direct download link: https://gitlab.com/tilasoldo/openfoa...tar?ref=master

Let me know if this works for you.

X.L.

Hi Tilasoldo,

Thank you!!! The code you provided functions almost perfectly and has been a great help for me. I've been so busy doing my impletation and running simulations I completely forgot to thank you for your effort. I made some minor changes to your code, so the freestream velocity can be read directly from a file.

So far I've been running Schubauer-Klebanoff and ERCOFTAC flat plate simulations, comparing the new gamma model and your implementation of gamma-Re_theta model against experimental results. Seems that the newer model is even more sensitive to low y+ values (y+<0.01) than the prior model. Low y+ values results slow convergence and moves the transition downstream during iterations. In addition, I am getting some weird results from certain tests where gamma-Re_theta model does not work as it should. Probably it is due to bad initial values, not sure though. Will keep you updated!

I attached some results from SK and T3A simulations if you are interested

Glad to help, your result are very interesting too. This model is indeed very demanding in terms of mesh at the wall… I didn't try myself with the T3A and SC models but they seem to be accurate enough compared to their cost in CPU time I guess ;)

Cheers!


Sent from my iPad using CFD Online Forum mobile app

incompatible dimensions for operation in compressible case
 

Hi, Sir

For compressible Case:

[26] --> FOAM FATAL ERROR:
[26] incompatible dimensions for operation
[intermittency[1 -3 -1 0 0 0 0] ] - [intermittency[0 0 -1 0 0 0 0] ]

```
// Re_theta equation
tmp<fvScalarMatrix> ReEqn
(
fvm::ddt(alpha, rho, Ret_)
+ fvm::div(alphaRhoPhi, Ret_)
- fvm::Sp(fvc::div(this->phi_), Ret_)
==
Ptheta*Reo_
- fvm::Sp(Ptheta, Ret_)
);

```

I change it to:

```
// Re_theta equation
tmp<fvScalarMatrix> ReEqn
(
fvm::ddt(alpha, rho, Ret_)
+ fvm::div(alphaRhoPhi, Ret_)
- fvm::Sp(fvc::div(alphaRhoPhi), Ret_)
==
Ptheta*Reo_*alpha*rho
- fvm::Sp(Ptheta*alpha*rho, Ret_)
);
```
It running..
Not sure it is corret or not?