CFD Online Discussion Forums - No more dynSmagorinsky in v. 2.3.0 ?

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)

- OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)

- - No more dynSmagorinsky in v. 2.3.0 ? (https://www.cfd-online.com/Forums/openfoam-solving/130957-no-more-dynsmagorinsky-v-2-3-0-a.html)

No more dynSmagorinsky in v. 2.3.0 ?

Hi everybody,
I am beginning on OpenFoam (using 2.1.0) and as I am trying to use dynamic smagorinsky model, I have an error message saying the following :

--> FOAM FATAL ERROR:
Unknown LESModel type dynSmagorinsky
Valid LESModel types:
15
(
DeardorffDiffStress
LRRDiffStress
Smagorinsky
SpalartAllmaras
SpalartAllmarasDDES
SpalartAllmarasIDDES
dynLagrangian
dynOneEqEddy
homogeneousDynOneEqEddy
homogeneousDynSmagorinsky
kOmegaSSTSAS
laminar
mixedSmagorinsky
oneEqEddy
spectEddyVisc
)

It seems many models that are supposed to be able available are missing.
Can someone help me with this ?

Thanks a lot !

No more dynSmagorinsky in v. 2.3.0 ?

Hi, I am new to OpenFoam and I want to use it for LES. I have seen that in previous versions a dynamic Smagorinsky turbulence model was in library, but it seems it has disappeared in v.2.3.0.
Does someone know solmething about it ?
Thanks a lot !

Greetings fbar,

I've merged your two threads on this topic, since they were related, although one referring to 2.1.0 and the other to 2.3.0.
The model was renamed back in OpenFOAM 1.7.x: https://github.com/OpenCFD/OpenFOAM-...890d543f61531d
Quoting the commit message:

Quote:

Renamed dynSmagorinsky to homogeneousDynSmagorinsky to avoid further confusion concerning the averaging used in this model.

Removed dynMixedSmagorinsky.

Best regards,
Bruno

Thanks a lot for your answer, I moved on homogeneousDynSmag but was not absolutely sure !

Hi,

== 1 ==
Is there any reason why 'dynMixedSmagorinsky' is removed from the recent versions?

Many thanks in advance.

Quote:

Originally Posted by HakikiCanakkaleli (Post 502660)

Is there any reason why 'dynMixedSmagorinsky' is removed from the recent versions?

Quick answer: From what I could see, it was renamed as well. Currently it's named "mixedSmagorinsky" and is located at "$FOAM_SRC/turbulenceModels/incompressible/LES/" in OpenFOAM 2.3.

Hi,

== 1 ==
Many thanks for your answer.

== 2 ==
I personally do not agree with this conclusion as 'mixedSmagorinsky' is not a dynamic model; but, a linear combination of scale-similarity and Smagorinsky(ish) models. Also, 'mixedSmagorinsky' class is present from the get-go.

Hi HakikiCanakkaleli,

Quote:

Originally Posted by HakikiCanakkaleli (Post 503273)

I personally do not agree with this conclusion as 'mixedSmagorinsky' is not a dynamic model; but, a linear combination of scale-similarity and Smagorinsky(ish) models. Also, 'mixedSmagorinsky' class is present from the get-go.

I unfortunately only took a quick glance at the description in the header files, and assumed they were the same thing.

Mmm... the only difference in the header for the dyn version was this:

Quote:

This version implements filtering to evaluate the coefficients in the model.

Interesting... the only other major difference is essentially this:

Code:

-class mixedSmagorinsky

+class dynMixedSmagorinsky

 :

     public scaleSimilarity,

-    public Smagorinsky

+    public dynSmagorinsky

In other words, the class "dynMixedSmagorinsky" would derive from "dynSmagorinsky" instead of just "Smagorinsky".

As far as I can see, the class "homogeneousDynSmagorinsky" which was originally "dynSmagorinsky", is based on code that has been in FOAM at least since 1991, if the copyright notice on the header is correct (have a look into the commit I mentioned before). The class seems to have been renamed for clarity, because there seems to be more than one way to implement a dynamic version of this model.
If you search online for "dynSmagorinsky", you'll see that other people have had to make some changes to that class/model, for example: http://powerlab.fsb.hr/ped/kturbo/Op...UniRostock.pdf

This leads me to deduce/think that when they renamed the class and added some more details, it was likely because someone reported this as a bug... :eek: and I found said bug just now: http://www.openfoam.org/mantisbt/view.php?id=99 - the bug report is pretty detailed, but there is no reference to the mixed variant.
From what I can deduce, the issue is that "dynMixedSmagorinsky" would no longer make much sense as "homogeneousDynMixedSmagorinsky", since they did not have the time to do more tests and validations with it. Or because in light of the topics discussed on the report, this modelling strategy would no longer make much sense.

If you want to know more details on what really happened, try using the contact form at www.openfoam.com ;)

Best regards,
Bruno