[braennstroem]

Did you have any success yet? I am actually
looking for it for a while now, but could not get it.

Did you try to include different libraries in the options, e.g.:

-I$(LIB_SRC)/LESmodels/LESdeltas/lnInclude \
-I$(LIB_SRC)/LESmodels/LESfilters/lnInclude \
-I$(LIB_SRC)/LESmodels \
-I$(LIB_SRC)/transportModels \
-I$(LIB_SRC)/LESmodels/LESmodel/lnInclude

-lincompressibleLESmodels \
-lincompressibleTransportModels \
-lLESdeltas \
-lLESfilters\

I am not sure, if it works!? Would be nice, if you have any success

Regards!
Fabian

[eugene]

First off, I can think of no reason why a fixedHeatFlux boundary should be templated. Well, I guess I could if I tried very hard, but you should not be using a template in this instance.

To use the compressible LES model library, you would have to include it in the options file:

-lcompressibleLESmodels

And to use that you would need some thermo libraries.

The problem is that you compile the finiteVolume library before the thermo library, so it is not a good idea to put fixedHeatFlux inside finiteVolume with all the other BCs. Cyclic dependencies are BAD. In addition, fixedHeatFlux cant go inside the thermo library, since the thermo library is compiled before the turbulence libraries.

So where to put it? My solution is to make a new library called fvPatchFields and to stick all the complicated multi-dependency BCs in there. You can then make the library available to applications by using the "libs" command in controlDict. The coodles options file is a good place to start for this BC.

[eugene]

Just to let you know, I have not built a fixedHeatFlux boundary of my own, so any code I posted before was just off the cuff stuff. That said, it is likely that division by Cp is required.

I took a look at your code and although it will compile, it wont run unless you create a permanent field called alphaEff that is registered with the database.

You really need to look up the turbulence and thermodynamic models and ask them for alphaEff and Cp directly. However, this will create the dependency issues I mentioned before.

[meiring_beyers]

Thanks Eugene

my feedback on the subdivision by Cp merely meant as a QA note to make sure I check the equation formulation.

I'll investigate the vfPatchField and "libs"combination. Thanks for help

[braennstroem]

Hi Meiring,

this might be interesting for you:

http://openfoamwiki.net/index.php/Contrib_wallHeatFlux

http://openfoamwiki.net/index.php/Ma...onditionedRoom

Fabian

[ep4]

Hi all,

running a steady laminar simulation with buoyantSimpleFoam, i 'm trying to impose a fix flux on a boundary of my computational domain. I use the perfect gas model of OpenFoam.

Reding this thread, i'm a little bit confused about the value of the heated flux i impose. For me, the heat flux is k*snGrad(T) (thermal conductivity*normal gradient). Here is mentioned alphaEff (the effective thermal diffusivity, equivalent to the thermal diffusivity for a laminar case i suppose). Dividing it by Cp and density, it's equivalent to the thermal conductivity.

My question concerns the value of alpha (or alphaEff) for my laminar case (model:laminar, turbulence off). I would say alphaEff=viscosity/Pr (Pr=Prandtl number). Is it right?

Thank you

Eric

[ep4]

Hi again,

actually, my previous question is not the problem for me.
The case i'm interested in is a 2D cavity, with one inlet and two outlet. It's a laminar case (Re=100). Imposing a constant flux on the bottom wall, i would like to have the maximal dimensionless temperature value which match with a paper i have.
Dimensionless temperature is theta = (T - Tinlet)/(q H/k) H is the height of the cavity.

Imposing a big temperature gradient (100000), i only have a small maximal dimensionless temperature (0.014) while expecting 0.9883.

I have checked different things but can't found the error i'm making and why the temperature are very low in my cavity.

Could someone help me?

Thank you

Eric

[ep4]

Last thing, the velocity values are in good agreement with the paper. The problem is really only with the temperature.