Learning Forum

[Shubham Sathe]

Hello 

 

I have been trying to do a non thermal equilibrium simulation in fluent. My domain is just a 2d section of a pipe (a rectangular domain with 3 inches length* 0.5 inches height and I also added an entry and exit sections (one rectangle at inlet and other rectangle at outlet) as well just so the flow is fully developed. My boundary conditions are also simple : constant temperature at wall (top and bottom of rectangle) , velocity inlet and pressure outlet. For the poros zone which is the entire pipe section (center rectangle) I gave the porosity, permeability, form coefficient, heat transfer coefficient. 

 

So I tried using viscous laminar model first and then the viscous k-epsilon turbulence models just to see how they behave (my velocity is 23 m/s which is high for laminar but still should show some difference in temperature profiles right??). But when I plot the temperature profiles at the outlet, I see little to no difference at all in both the temperature profiles. I am attaching these temperature profiles in the attachment, (on X axis is the Y direction and Y axis is temperature), please have a look.  

 

Also I tried doing the same simulation with the same domain (rectangular with inlet and exit sections) without poros zone and when I use laminar and turbulence models at same velocity 23 m/s, the temperature profiles are as they were supposed to be for a pipe.

[Rob]

Please don't multipost, it's annoying and wastes my time cleaning up.

If you apply a resistance to flow what happens to the flow profile? 

[nandacd]

Hi Rob

Sorry for the duplicate post. 
I have given inertial resistance as an input to the porous domain. 

I cannot give any physical resistance like adding a geometry since I have none.

Thankyou for your reply.

[Rob]

Yes, so you're increasing the pressure loss as a function of velocity and velocity squared. What might that do to the velocity profile?

[nandacd]

Thankyou for your reply. Right we are increasing the pressure velocity squared, but still when solving the energy equations like laminar and k-epsiilon models, there should be some change in temperature profiles right??

The velocity profile might not be much different for laminar flow. But for turbulent flows it will increase mixing.

Also I forgot to add the temperature profile, which I am attaching in this reply. Please have a look.

 

Best

Nanda

 

[Rob]

Maybe. If I increase the pressure loss more for a region of higher velocity flow should I expect the flow profile to be retained? 

You may also want to look at mesh resolution - not just near wall. 

[nandacd]

The flow profile will be the same just that the maximum velocity will be decreased , since there is resistance.

My mesh has a maximum orthogonal quality of 0.56.

But the thing is in FLUENT manual , there is this mention about turbulence and laminar zone. I am pasting the same here in this reply for your reference. I was hoping whether there might be any bug or something , because the turbulence generated has no effect on my heat transfer. Since I am getting the same result for my simulation with and without any turbulence.

 

Treatment of Turbulence in Porous Media

---

ANSYS FLUENT will, by default, solve the standard conservation equations for turbulence quantities in the porous medium. In this default approach, turbulence in the medium is treated as though the solid medium has no effect on the turbulence generation or dissipation rates. This assumption may be reasonable if the medium's permeability is quite large and the geometric scale of the medium does not interact with the scale of the turbulent eddies. In other instances, however, you may want to suppress the effect of turbulence in the medium.

If you are using one of the turbulence models (with the exception of the Large Eddy Simulation (LES) model), you can suppress the effect of turbulence in a porous region by setting the turbulent contribution to viscosity, , equal to zero. When you choose this option, ANSYS FLUENT will transport the inlet turbulence quantities through the medium, but their effect on the fluid mixing and momentum will be ignored. In addition, the generation of turbulence will be set to zero in the medium. This modeling strategy is enabled by turning on the Laminar Zone option in the Fluid dialog box. Enabling this option implies that  is zero and that generation of turbulence will be zero in this porous zone. Disabling the option (the default) implies that turbulence will be computed in the porous region just as in the bulk fluid flow. Refer to Section  7.2.1 for details about using the Laminar Zone option

[Rob]

If the peak velocity is decreased by the porous zone, but the mass (volume) flow is unchanged where is the rest of the material? 

[nandacd]

Hi Rob

So I tried plotting the velocity profile, for both the models of laminar and k-epsilon. This is what I have gotten. I am attaching the velocity profile in this reply. So the issue here is the turbulence generated in fluent is not very significant that there is much difference in laminar and turbulent models. 

But I have also tried the same simulations without the porous medium and just a simple pipe flow. There I did not find any issues with the velocity and temperature profiles, they were as expected for laminar and turbulence model

 

[Rob]

You have a porous media - plot the velocity profile towards the end of that and compare the two lines. Then think about how many diameters you'd expect for a fully developed profile to require. 

Note, you may need to review the mesh resolution too - y+ is for the near wall cell, you may need more mesh in the bulk flow too.