[cavida]

Hi bros
How can I set the code or other settings to have a calculation of the fluid flow just in one direction like x. (unidirectional flow)

Note that my model is 3D and I have to solve 3 equations: fluid flow, electric field and magnetic field (my model is MHD)

I want to solve the electric and magnetic fileds in 3 directions (x,y,z) but the fluid flow just in x direction!

Is it possible?

[Saideep]

Hi,

I am not sure if i understood your question properly.

As you are using a 3D model all your equations and components of interest are computed in all 3 directions.

Now, by your inlet boundary condition for fluid velocity{considering this as your fluid flow}, we generally specify the value for velocity of injection as 3D. Here you can specify one direction flow either of x/y/z component.

Based on the equation you are solving in space the flow is propagated in 3 directions.
It would get complicated if you also want to modify your solver to analyze only 1 direction flow for a 3D case. {I guess also you are damping out the accuracy of your actual solution in such case}.

However if you are interested in only one set of results, you can resolve the component of fluid flow using foamCalc utility and extract essential data of interest.

Hope this helps;
Saideep

[cavida]

Thanks for responding dear Saideep

In fact I'm going to compare the results of considering 3-directional fluid flow with the results of unidirectional fluid flow, but the other fields must be solved in 3-directional.

I think foamCalc just extracts the results of 3D solution and that would not work for my comparison.

[blais.bruno]

Originally Posted by cavida

Hi bros
How can I set the code or other settings to have a calculation of the fluid flow just in one direction like x. (unidirectional flow)

Note that my model is 3D and I have to solve 3 equations: fluid flow, electric field and magnetic field (my model is MHD)

I want to solve the electric and magnetic fileds in 3 directions (x,y,z) but the fluid flow just in x direction!

Is it possible?

What you could do, I am not sure is multiply your flow equations by a [1 0 0]^T vector.

So this way for Ux you would have a normal navier stokes equation, but for the rest you would just have ddt=0 and this would lead to a trivial 0=0 solution. This is the easiest way I can think off to disable the other components while maintaining a 3D mesh... Someone else might have a better solution