CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (http://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Appropriate model for turbulent, steady state pipe elbow flow (http://www.cfd-online.com/Forums/openfoam-solving/62833-appropriate-model-turbulent-steady-state-pipe-elbow-flow.html)

milos March 21, 2009 10:22

Appropriate model for turbulent, steady state pipe elbow flow
 
2 Attachment(s)
Hi everyone!

Ok, here's the deal: I am investigating a case of turbulent, steady state pipe elbow flow or to be more precise, pressure drops due to pipe elbows (DN 50). Now, I believe that the best solver for this kind of thing is simpleFoam, but the thing is that I don't get the appropriate stream (velocity vector) vortices in the elbow. I do get the vortices, but they start appearing after the elbow, not in the elbow as the experience tells us.

I attached an example pic named U_vortices.jpg from one science projects I found done in Fluent. Notice that the text beneath the pic says it is elbow-MIDSECTION (45 deg or half-way through the elbow). I actually used very similar parameters of the flow as in that simulation, which includes: Re around 540000, with water as a fluid (nu at 20 deg C = 1.004e-6).

Now take a look my other attachment named Stream 1.jpg , which is what I got.

Where's the catch?

My pressure BC's are:


inlet - zeroGradient
outlet - fixedValue (0)
wall - zeroGradient

My velocity BC's are:

inlet - fixedValue (10 0 0)
outlet - zeroGradient
wall - fixedValue (0 0 0)

Thanx!

Milos

gskillas March 24, 2009 18:30

Make the images really comparable :-)
 
Hello Milos

could you also make a plot similar to U_vortices.jpg? I find it very hard to compare your data with the Fluent simulation... I even suspect that they might be closer than youŽd think at first glance.

Regards,

George

milos March 25, 2009 10:08

1 Attachment(s)
Thnx for the reply George. :)

After several hours of roaming through ParaView I found the option that provided me with the speed intensity scalars (namely: Ux, Uy, Uz), which in addition gave me the attached photo. I guess I was really close all the time... :)

The thing is just that the guy, who has done the example simulation in Fluent, actually drew, as it states beneath his figure - IN-PLANE velocity VECTORS. I'm not sure whether I can do a similar thing in ParaView... Does anyone know how to do that in ParaView? The closest I got was graphing the speed component intensity as a scalar.

kjetil July 7, 2009 11:09

Quote:

Originally Posted by milos (Post 210225)
Hi everyone!


My pressure BC's are:


inlet - zeroGradient
outlet - fixedValue (0)
wall - zeroGradient

My velocity BC's are:

inlet - fixedValue (10 0 0)
outlet - zeroGradient
wall - fixedValue (0 0 0)

Thanx!

Milos


I'm on a quest for pipe simulation myself - and would like to know what files you keep in your '0' directory for this kind of simulation? The tutorials I have been looking through all seem to use gases - and I wonder if I can simplify the setup using a fluid (water)? The problem is that my simulation using simpleFoam is very very slow, basing it upon a previous gas-case tutorial.

milos July 9, 2009 02:24

I used simpleFoam solver, since it is for incompressible flows... I simply used the files from the tutorial folder.


All times are GMT -4. The time now is 11:59.