Dimensions in ScalarTransportFoam

Gesetzt · November 26, 2017, 10:15

Hi,

I have an simple question. I'm referring to OF 4.1.

There is the definition of the equation:
fvScalarMatrix TEqn
(
fvm::ddt(T)
+ fvm::div(phi, T)
- fvm::laplacian(DT, T)
==
fvOptions(T)
);

So, this means: $\frac{\partial T}{\partial t} +\nabla\cdot{(\phi T)}-\nabla(\Gamma \nabla T)=0$ .

For incompressible fluids phi means: $U \cdot A$ . I don't get the dimensions right here. For the first term it's obviously $\frac{K}{s}$ . But for the second term phi has to be U, to get the right dimension.

I looked up phi and it says: surfaceScalarField phi

(
IOobject
(
"phi",
runTime.timeName(),
mesh,
IOobject::READ_IF_PRESENT,
IOobject::AUTO_WRITE
),
fvc::flux(U)
);

So what exactly is meant with flux(U), is this without the area? I couldn't find the code definition of flux(U).

I hope someone can help me to clarify this topic.

Best regards

Gesetzt

Traction · November 27, 2017, 04:55

Quote:

Originally Posted by Gesetzt

So what exactly is meant with flux(U), is this without the area? I couldn't find the code definition of flux(U).

Hi,
please have a look at src/finiteVolume/finiteVolume/fvc/fvcFlux.C where the volVectorField is U in your case! Makes things clearer ?

Gesetzt · December 3, 2017, 06:48

Hey,

Thanks. I think it's clear now. Part of the problem was, that I didn't thought about the fact we are using FVM, so for the discretisation over all faces we need the area.

Best Regards,

Gesetzt

svramana · February 16, 2018, 03:27

Quote:

Originally Posted by Gesetzt

Hi,

I have an simple question. I'm referring to OF 4.1.

There is the definition of the equation:
fvScalarMatrix TEqn
(
fvm::ddt(T)
+ fvm::div(phi, T)
- fvm::laplacian(DT, T)
==
fvOptions(T)
);

So, this means: $\frac{\partial T}{\partial t} +\nabla\cdot{(\phi T)}-\nabla(\Gamma \nabla T)=0$ .

For incompressible fluids phi means: $U \cdot A$ . I don't get the dimensions right here. For the first term it's obviously $\frac{K}{s}$ . But for the second term phi has to be U, to get the right dimension.

I looked up phi and it says: surfaceScalarField phi

(
IOobject
(
"phi",
runTime.timeName(),
mesh,
IOobject::READ_IF_PRESENT,
IOobject::AUTO_WRITE
),
fvc::flux(U)
);

So what exactly is meant with flux(U), is this without the area? I couldn't find the code definition of flux(U).

I hope someone can help me to clarify this topic.

Best regards

Gesetzt

HI,
i have question about TEqn. can this be valid for porous simple foam also.

November 26, 2017, 10:15	Dimensions in ScalarTransportFoam	#1
Gesetzt New Member Dan Join Date: Sep 2016 Posts: 20 Rep Power: 10	Hi, I have an simple question. I'm referring to OF 4.1. There is the definition of the equation: fvScalarMatrix TEqn ( fvm::ddt(T) + fvm::div(phi, T) - fvm::laplacian(DT, T) == fvOptions(T) ); So, this means: $\frac{\partial T}{\partial t} +\nabla\cdot{(\phi T)}-\nabla(\Gamma \nabla T)=0$ . For incompressible fluids phi means: $U \cdot A$ . I don't get the dimensions right here. For the first term it's obviously $\frac{K}{s}$ . But for the second term phi has to be U, to get the right dimension. I looked up phi and it says: surfaceScalarField phi ( IOobject ( "phi", runTime.timeName(), mesh, IOobject::READ_IF_PRESENT, IOobject::AUTO_WRITE ), fvc::flux(U) ); So what exactly is meant with flux(U), is this without the area? I couldn't find the code definition of flux(U). I hope someone can help me to clarify this topic. Best regards Gesetzt

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Different dimensions for = dimensions : [0 2 -1 0 0 0 0] = [0 4 -3 0 0 0 0]	mohakbl111	OpenFOAM Running, Solving & CFD	5	February 6, 2019 06:37
FOAM FATAL ERROR for Different dimensions	rupesh_w	OpenFOAM Running, Solving & CFD	2	April 27, 2016 05:14
Incompatible dimensions for operation	ruben23	OpenFOAM Running, Solving & CFD	2	June 12, 2015 05:14
Software that can stretch the overall dimensions of a mesh.	Robert_B	Main CFD Forum	2	January 27, 2012 08:39
Units in scalartransportfoam	Frithjof	OpenFOAM	1	January 5, 2011 11:41

December 3, 2017, 06:48		#3
Gesetzt New Member Dan Join Date: Sep 2016 Posts: 20 Rep Power: 10	Hey, Thanks. I think it's clear now. Part of the problem was, that I didn't thought about the fact we are using FVM, so for the discretisation over all faces we need the area. Best Regards, Gesetzt