Source to drive velocity to zero

philippose · August 31, 2017, 06:54

Hello everybody,
A Good day to you all.

I have a quick question which I have been breaking my head over for a while now.

I need to simulate a case where depending on the flow through a patch, I need to increase or decrease the flow resistance at another location within the mesh, something like a variable orifice.

My intention is to do this without mesh morphing or topological changes, but rather, to use something like an "alpha" field to somehow drive the velocity at the right locations in the mesh to zero to emulate a blockage.

As of now, I have included the following source term into the Ueqn of a modified simpleFoam solver...:

Code:

    tmp<fvVectorMatrix> tUEqn
    (
        fvm::div(phi, U)
      + MRF.DDt(U)
      + turbulence->divDevReff(U)
      + fvm::Sp(alpha*max(mag(U)).value(), U)
     ==
        fvOptions(U)
    );

The part with the "fvm::Sp" is my additional source. My alpha field is a digital field.. either 0 or 1000.

The part:

Code:

 max(mag(U)).value()

Is only to get a resulting value which is dependent on the velocity.

The code seems to be working, but I dont see the velocity being driven down to zero.... it creates a kind of blockage, but I have had cases where even though the entire channel is blocked (alpha = 1000), I still have flow through, though, lesser than if there was no block.

However, my intention is to drive the velocity in all the cells where the alpha field is 1000 to zero.

Could someone please help me out?

Thank you.

Wish you a good afternoon ahead.

Regards,
Philippose

philippose · September 1, 2017, 04:53

Hello again,
A Good Friday Morning to everyone :-)

So, I have basically solved the problem I had with injecting a source into the simulation to drive the velocity to zero at various locations.

It was a conceptual oversight on my part. It would be too involved to explain it here without going into the brackground, which would make this post too long and boring.

If anyone is interested, you can let me know, and I can try to explain the problem, and the potential solution.

Wish you all a good weekend ahead.

Regards,
Philippose

krikre · October 28, 2017, 12:17

Hey!

I'm curious how you did it. I need simple solidification.

My solution was to iteratively correct the body force $\textbf{b}$ in the pimple loop.
$\textbf{b}_{n+1} =\textbf{b}_n - r \, \textbf{U}_n,$
where

in $[(\mathrm{N}/\mathrm{m}^3)/(\mathrm{m}/\mathrm{s})]$ is a relaxation factor.

But I don't think it's working right.

August 31, 2017, 06:54	Source to drive velocity to zero	#1
philippose Senior Member Philippose Rajan Join Date: Mar 2009 Location: Germany Posts: 552 Rep Power: 25	Hello everybody, A Good day to you all. I have a quick question which I have been breaking my head over for a while now. I need to simulate a case where depending on the flow through a patch, I need to increase or decrease the flow resistance at another location within the mesh, something like a variable orifice. My intention is to do this without mesh morphing or topological changes, but rather, to use something like an "alpha" field to somehow drive the velocity at the right locations in the mesh to zero to emulate a blockage. As of now, I have included the following source term into the Ueqn of a modified simpleFoam solver...: Code: tmp<fvVectorMatrix> tUEqn ( fvm::div(phi, U) + MRF.DDt(U) + turbulence->divDevReff(U) + fvm::Sp(alpha*max(mag(U)).value(), U) == fvOptions(U) ); The part with the "fvm::Sp" is my additional source. My alpha field is a digital field.. either 0 or 1000. The part: Code: max(mag(U)).value() Is only to get a resulting value which is dependent on the velocity. The code seems to be working, but I dont see the velocity being driven down to zero.... it creates a kind of blockage, but I have had cases where even though the entire channel is blocked (alpha = 1000), I still have flow through, though, lesser than if there was no block. However, my intention is to drive the velocity in all the cells where the alpha field is 1000 to zero. Could someone please help me out? Thank you. Wish you a good afternoon ahead. Regards, Philippose

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[swak4Foam] swak4foam building problem	GGerber	OpenFOAM Community Contributions	54	April 24, 2015 16:02
SparceImage v1.7.x Issue on MAC OS X	rcarmi	OpenFOAM Installation	4	August 14, 2014 06:42
Problem compiling a custom Lagrangian library	brbbhatti	OpenFOAM Programming & Development	2	July 7, 2014 11:32
Trouble compiling utilities using source-built OpenFOAM	Artur	OpenFOAM Programming & Development	14	October 29, 2013 10:59
[swak4Foam] funkySetFields compilation error	tayo	OpenFOAM Community Contributions	39	December 3, 2012 05:18

September 1, 2017, 04:53		#2
philippose Senior Member Philippose Rajan Join Date: Mar 2009 Location: Germany Posts: 552 Rep Power: 25	Hello again, A Good Friday Morning to everyone :-) So, I have basically solved the problem I had with injecting a source into the simulation to drive the velocity to zero at various locations. It was a conceptual oversight on my part. It would be too involved to explain it here without going into the brackground, which would make this post too long and boring. If anyone is interested, you can let me know, and I can try to explain the problem, and the potential solution. Wish you all a good weekend ahead. Regards, Philippose

October 28, 2017, 12:17		#3
krikre Member Kristjan Join Date: Apr 2017 Location: Slovenia Posts: 36 Rep Power: 9	Hey! I'm curious how you did it. I need simple solidification. My solution was to iteratively correct the body force $\textbf{b}$ in the pimple loop. $\textbf{b}_{n+1} =\textbf{b}_n - r \, \textbf{U}_n,$ where in $[(\mathrm{N}/\mathrm{m}^3)/(\mathrm{m}/\mathrm{s})]$ is a relaxation factor. But I don't think it's working right.