[Vernon-CANADA]

Dear all,

I'm dealing with a cfd problem and i can't find a solution so I need help.
I have a pipe in which I know the flow rate that I insert, and that's all.
Outlet, water goes in another part and I don't know the pressure.

So my flow is incompressible and stationary with k-w model since it's a rotating machine.
But I want to study pressure inside the pipe but I don't know the outlet pressure.

I tried velocity inlet and outlet with flow rate conservation but it doesn't converge.

Does anyone know what to do please?

[FMDenaro]

Originally Posted by Vernon-CANADA

Dear all,

I'm dealing with a cfd problem and i can't find a solution so I need help.
I have a pipe in which I know the flow rate that I insert, and that's all.
Outlet, water goes in another part and I don't know the pressure.

So my flow is incompressible and stationary with k-w model since it's a rotating machine.
But I want to study pressure inside the pipe but I don't know the outlet pressure.

I tried velocity inlet and outlet with flow rate conservation but it doesn't converge.

Does anyone know what to do please?

If you are solving the incompressible flow formulation, the mass conservation is simply div v =0. The pressure equation is derived from that, there is no thermodynamic law.
In such a case, the inflow with Dirichlet and the outflow with Neumann BCs for the velocity will work. The pressure boundary conditions for the Poiisson equation are derived from the velocity BCs. You can fix arbitrary a constant value for the pressure in a point.

[Vernon-CANADA]

Yeah I've just tried to apply an arbitrary pressure outlet.
K-w model crashes, k-epsilon gives me negative pressures inside the pipe, what is impossible.
Did you encounter that kind of problem before?

[FMDenaro]

Originally Posted by Vernon-CANADA

Yeah I've just tried to apply an arbitrary pressure outlet.
K-w model crashes, k-epsilon gives me negative pressures inside the pipe, what is impossible.
Did you encounter that kind of problem before?

Have you checked your setting in a model case without any turbulence model? What kind of BC for the velocity in outflow?

[Vernon-CANADA]

Originally Posted by FMDenaro

Have you checked your setting in a model case without any turbulence model? What kind of BC for the velocity in outflow?

It's a rotating pipe, so volumic loads, monophasic water, inlet the velocity and outlet an arbitrary pressure but i have negative pressures.
I run without any turbulence to see.

[FMDenaro]

Originally Posted by Vernon-CANADA

It's a rotating pipe, so volumic loads, monophasic water, inlet the velocity and outlet an arbitrary pressure but i have negative pressures.
I run without any turbulence to see.

Ok, let's first check the setting for a steady laminar flow at low Re number. If you have still problems you are setting wrong outflow condition.

What software are you using?

[Vernon-CANADA]

Originally Posted by FMDenaro

Ok, let's first check the setting for a steady laminar flow at low Re number. If you have still problems you are setting wrong outflow condition.

What software are you using?

I'm using comsol multiphysics and in my previous company I was using Anys so i'm a beginner with Comsol.
The computation is running, laminar with low reynolds

[FMDenaro]

Originally Posted by Vernon-CANADA

I'm using comsol multiphysics and in my previous company I was using Anys so i'm a beginner with Comsol.
The computation is running, laminar with low reynolds

Ok, check if you have a fully convergence and if mass conservation is ensured with your setting. If the solution appears suitable, the problem you have is with the turbulence model and its required BCs

[Vernon-CANADA]

Originally Posted by FMDenaro

Ok, check if you have a fully convergence and if mass conservation is ensured with your setting. If the solution appears suitable, the problem you have is with the turbulence model and its required BCs

The laminar computation is over and i have an error message: "Undefined Value Found In The Equation Residual Vector"
So I have used coarse mesh and ran other analysis just to see the impact of my boundary conditions by modying.
I have another message: "failed to find a solution. Maximum number of Newton iterations reached. Returned solution is not converged"

I don't understand the problem.
It's a pipe, water goes throught and it's two parts. One fixed and another rotating.

I apply volumic load for the rotation, I apply no slip on walls, the a velocity inlet and pressure outlet.
where is the problem?

[FMDenaro]

I don’t use Comsol and that is a my limit, but at this step I would check a very simple test case like the laminar solution in a pipe with a parabolic inlet velocity and convective bcs

[FMDenaro]

I don’t use Comsol and that is a my limit, but at this step I would check a very simple test case like the laminar solution in a pipe with a parabolic inlet velocity and convective bcs

[Vernon-CANADA]

Originally Posted by FMDenaro

I don’t use Comsol and that is a my limit, but at this step I would check a very simple test case like the laminar solution in a pipe with a parabolic inlet velocity and convective bcs

I have run the simple laminar study and it converges in 39s when i disable coriolis volumic loads from rotation of the pipe!

I'm crazy lol