Solving stationary two-phase Laplace-equation

Martin123 · June 3, 2009, 19:39

Sorry first for my beginner question, I am quite new to CFD. I would like to solve the Laplace equation for inhomogeneous media:

k div grad U + grad k grad U = 0

For starting, I tried to solve the first term only and managed to setup the k-field (volscalarfield) on run-time. However, I don't see any influence of varying k when using fvm:: discretization - I solved

fvm::laplacian(k, U)

Am I on the right way or should I rethink completely my approach? I am very unsure how OF deals with the extra volScalarField from K: I want to solve certainly for U but I wonder what OF would see with my equation above? Should I better use somehow dictionary values for k and grad k?

Martin123 · June 5, 2009, 12:51

OMG I am really lost. What I tried:

...
volVectorField gradepsilon = fvc::grad(epsilon);
volVectorField gradU = fvc::grad(U);

solve
(
fvm::laplacian(U) + ((gradepsilon & gradU)/epsilon)
);

But still, I don't get correct results. Is my assumption right, that I may not use a second geometric field here inside since it is considered as an additional degree of freedom? Did I understand correctly that: if I use fvc:: this is added as a source term whereas all fvm:: operations cause the DOF (e.g. U) to be considered?

Is it alternatively possible to construct a dictionary from a volScalarField? Or taking dictionary values from a run-time function?

Many thanks in advance for your help.

June 3, 2009, 19:39	Solving stationary two-phase Laplace-equation	#1
Martin123 New Member Martin Tek Join Date: May 2009 Posts: 7 Rep Power: 17	Sorry first for my beginner question, I am quite new to CFD. I would like to solve the Laplace equation for inhomogeneous media: k div grad U + grad k grad U = 0 For starting, I tried to solve the first term only and managed to setup the k-field (volscalarfield) on run-time. However, I don't see any influence of varying k when using fvm:: discretization - I solved fvm::laplacian(k, U) Am I on the right way or should I rethink completely my approach? I am very unsure how OF deals with the extra volScalarField from K: I want to solve certainly for U but I wonder what OF would see with my equation above? Should I better use somehow dictionary values for k and grad k?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Moving mesh	Niklas Wikstrom (Wikstrom)	OpenFOAM Running, Solving & CFD	122	June 15, 2014 07:20
How to write k and epsilon before the abnormal end	xiuying	OpenFOAM Running, Solving & CFD	8	August 27, 2013 16:33
Differences between serial and parallel runs	carsten	OpenFOAM Bugs	11	September 12, 2008 12:16
IcoFoam parallel woes	msrinath80	OpenFOAM Running, Solving & CFD	9	July 22, 2007 03:58
Could anybody help me see this error and give help	liugx212	OpenFOAM Running, Solving & CFD	3	January 4, 2006 19:07

June 5, 2009, 12:51		#2
Martin123 New Member Martin Tek Join Date: May 2009 Posts: 7 Rep Power: 17	OMG I am really lost. What I tried: ... volVectorField gradepsilon = fvc::grad(epsilon); volVectorField gradU = fvc::grad(U); solve ( fvm::laplacian(U) + ((gradepsilon & gradU)/epsilon) ); But still, I don't get correct results. Is my assumption right, that I may not use a second geometric field here inside since it is considered as an additional degree of freedom? Did I understand correctly that: if I use fvc:: this is added as a source term whereas all fvm:: operations cause the DOF (e.g. U) to be considered? Is it alternatively possible to construct a dictionary from a volScalarField? Or taking dictionary values from a run-time function? Many thanks in advance for your help.