turbulence model in fire simulation

[xiaojun218]

Hi, all
Is k-ωSST two-equation model suitable to modeling the
turbulence in the fire simulaton? I want to replace the LES model with
k-ωSST two-equation model in FDS, but I am not very sure wether k-
ωSST model can do a better job than k-ε model or in the simulaton.
Any reply will be appreciated.

Regards

[rmcdermo]

k-epsilon and k-omega are typically used in RANS. FDS is an LES
code. I would not waste time on k-omega. You can set DYNSMAG=T on
MISC to invoke the dynamic Smagorinsky model if the default constant
coefficient Smagorinsky model is insufficient.

[Fabian Braennstroem]

Hi,

a hybrid RANS-LES komega model could be quite good approach to get good
wall behavior.

Regards!
Fabian

> --~--~---------~--~----~------------~-------~--~----~
> You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
> To post to this group, send email to fds...@googlegroups.com

[rmcdermo]

Fabian,

Wouldn't you agree that such an approach is completely flow
dependent? And dependent on what you mean by "wall behavior"? Hybrid
methods have been around for years (decades), but these all fail for
detached flows. Then you get into the whole arena of DES (Detached
Eddy Simulations). We have found that the Werner Wengle wall model is
a good compromise between sophistication and simplicity.

Randy

[Fabian Braennstroem]

Hello Randy,

On 10/13/2009 08:28 PM, rmcdermo wrote:
> Fabian,
>
> Wouldn't you agree that such an approach is completely flow
> dependent? And dependent on what you mean by "wall behavior"?

With wall behavior I actually meant, that is not too easy to get a well
resolved LES on the wall and that a hybrid RANS-LES model, which could
be a classical DES model or any other kind of hybrid RANS-LES model,
could improve the predictions e.g. of heat transfer at the wall.

Regards!
Fabian

[Rahman]

Dear Randy,

I want  simulation the ventilation system  of Amphitheatre.
for this work rans turbulence model is Suitable but FDS is a LES code.
For this reason I wanted to know which of turbulence models of LES is suitable for coarse mesh?

[Randy McDermott]

We do not recommend playing with the turbulence models in FDS.  Changing anything from the default values basically means you are running a code that has not been validated.  The models that have been selected as defaults have performed the best for our documented tests over a range of conditions, and I suspect you will find our validation cases are run at sufficiently coarse grid resolution.

--

You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.

To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To post to this group, send email to fds...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/98952a62-ecdb-44aa-88e7-b066882f36a7%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.