[wildfire230]

Hi All,


I have a problem where I have a volTensorField that is a stress tensor (if you are curious, it is the Maxwell stress tensor), and I am trying to use this stress tensor to calculate the total force on a patch.


I have been trying to use a surfaceFieldValue entry in my controlDict file that looks like this:


maxwellForcesSphere
{
type surfaceFieldValue;
libs ("libfieldFunctionObjects.so");

log true;
writeControl timeStep;
writeInterval 1;
writeFields false;

surfaceFormat none;

regionType patch;
name sphere;

operation areaNormalIntegrate;

fields
(
maxwellStressTensor
);
}


Essentially, the force calculation should work like this: take the maxwellStressTensor, which is already calculated, take the inner product of this with the surface normal direction on the patch, and integrate this over the patch. It's my understanding that this is what 'areaNormalIntegrate' should be doing, but it's very possible I am misunderstanding this.


Could someone please advise me on what I am doing wrong? When I run this as is it gives an output like this:


areaNormalIntegrate(sphere) of maxwellStressTensor = (0 0 0 0 0 0 0 0 0)


I am quite confident this should not be zero, and also the output should be a vector rather than a tensor, due to the inner product with the surface normal direction.


Thanks for your help


EDIT: I think another way to frame this question would be: suppose you calculate the velocity gradients tensor in your flow solver. Is it possible to calculate the total viscous force on a patch using surfaceFieldValue?