[su_junwei]

Dear foamers

what is the difference between fvc::surfaceIntegrate and fvc::surfaceSum?

Isn't it an additional division of volume in fvc::surfaceIntegrate ?

I have just noticed that they are different when suming the surface field

In fvc::surfaceIntegrate, it is
forAll(owner, facei)
{
ivf[owner[facei]] += issf[facei];
ivf[neighbour[facei]] -= issf[facei];
}
while in fvc::surfaceSum, it is

forAll(owner, facei)
{
ivf[owner[facei]] += issf[facei];
ivf[neighbour[facei]] += issf[facei];
}
Why they are different when dealing with neighbour cell for the current face?

Is it a bug for fvc::surfaceIntegrate?

Junwei

[henry]

"surfaceIntegrate" forms the basis of the Gauss integrals for evaluating explicit derivatives, see e.g. fvc::div. The reason there is a change of sign is that the face-areas are outward-pointing from the "owner" and hence inward-pointing to the "neighbour". This is assuming what is being "summed" is some kind of flux, i.e. something multiplied or dotted with the cell face areas.

"surfaceSum" is simply that, a sum of a surface field over the cell-faces which is needed for some forms of averaging.

H

[nimasam]

hi dear friend

what dose "fvc :: reconstruct" do exactly ?

i have a "surfaceScalarField" and i want to change it to a "volumeScalarField", what can i do ?
can i use fvc :: reconstruct ? whats its effect on data?

[Anne Lincke]

The surfaceIntegrate utility sounds very useful to me.
Is it possible to apply this for a face-based source term inside the UEqn()?

For which objects can we use this method?

When applying this to velocity U I obtain an error message.

Thank you for an answer in advance.

Kind Regards
Anne Lincke

[wyldckat]

Greetings Anne,

It would help if you provided more information about the problem at hand.

I can only guess that the following might help you understand better how OpenFOAM's "reconstruct" and "interpolate" work: https://github.com/wyldckat/reconstr...te-fields/wiki

Best regards,
Bruno