[paul6030]

I'm currently working on a project in ANSYS fluent designing a force air-cooled oil tank by cooling tubes. I'm also very new at dealing with fluent so I don't know much on how to set up the solution.

The tank must be full before it starts draining.
Here are some specs that I'm not sure how to set up

Oil Inlet temperature = 225˚F
Oil Inlet Pressure = 60psi
Air pressure = .008373 psi
Air temperature = 95˚F
Air density = .0614 lbm/ft3

[lucar4980]

It is not clear which is your problem. If you have the boundary conditions, you just have to properly set them.
If the air speed is low, you can simulate both air and oil as incompressible fluids. In addition to the BC you mentioned, you have to set the mass flow or the outlet static pressure.
Hope this helps.

[paul6030]

I think the BC's and Cell Zone condition might be my problem then. I'm not sure in setting them up.

I have figured out the inlet BC as a pressure inlet at 60 psi and I know the outlet is supposed to be set as a pressure outlet, but I'm not sure what the psi is supposed to be for the outlet?
Nor do I know what type of inlet my air-flow is supposed to be.
Lastly, how do I have the tank fill up before it starts flowing through the outlet?

[ghost82]

Hi Paul,
I suggest to look at tutorials before starting your case.
Boundary conditions are inputs, if you don't know your inputs you cannot calculate anything and you can't ask here what are your boundary conditions, simply because we don't know .
If you want to set a pressure outlet bc you must know what is the pressure outlet, and what are the flow conditions at the outlet.
Air inlet bc depends on your case: for example if air enters at 60 psi and the outlet is at atmospheric pressure you have to treat air as compressible, so you can set a mass flow inlet bc.
If air flow is incompressible (i.e. density does not change too much in your domain) you can set a pressure inlet or velocity inlet bc.
It seems you need to set up a transient vof simulation, so you can initialize your domain filled with primary phase and then patch the zone you want with the secondary phase.

[lucar4980]

Well, let me summarize your problem for what I have understood: you have a tank full of hot oil. There are also some pipes full of cold air submerged in the oil. The two fluids are always separated. The pipes walls are "coupled" in order to allow heat transfer.
You have the following BC: air inlet, air outlet, oil inlet, oil outlet.
You have to enable energy equation in order to simulate heat transfer, and also to properly set turbulence model (check Re number if it is necessary).

If the fluids are incompressible (low speed for air) you can set outflow BC (but you need to set also the mass flow at inlet or velocity). Otherwise, set the pressure Drop and check for the results.

[paul6030]

Yes Lucar, that is exactly the problem. I'll see if this works.

[paul6030]

I have the tube walls coupled, but I'm getting an error in my solution.

Error: Divergence detected in AMG solver: pressure correction
Error Object: #f

[lucar4980]

You have to change the initialization: avoid strong gradients. Try also to reduce Under Relaxation Factors.

Check this post: http://www.cfd-online.com/Forums/flu...omentum-2.html

[paul6030]

Okay that problem is fixed. Only thing is to specify which liquid/gas goes in which opening. I've tried to use Mixture and Volume of liquid, but that didn't help with 2 phases and I'm just not getting the right result for flow.

[lucar4980]

You have to enable "models->species->species transport".
The you have to set your mixture (Air + Oil) in the material menu.
In this way, you will be able to set a "concentration" for each BC. For Air you will set 100% air, 0% oil. For oil 100%oil 0%air.

Consider that the last specie you set in the mixture is used as "complement to 1", so it is automatically set at the BC simply as "1 - other species".
This is the reason why you will be able to set just one single number at each BC.
Hope this helps.

[paul6030]

Does the model "multiphase" need to be on?

[lucar4980]

No, I don't think you need multiphase flow.

[paul6030]

In the materials section do I need to specify it as a mixture?

[lucar4980]

Yes, you have to set it as a mixture even though you will utilize both fluids as pure species. I don't think there exist other options...it should work.

[paul6030]

Everything seems right, but I'm getting an error.

"Error: Divergence detected in AMG solver: pressure correction.

[lucar4980]

Originally Posted by lucar4980

You have to change the initialization: avoid strong gradients. Try also to reduce Under Relaxation Factors.

Check this post: http://www.cfd-online.com/Forums/flu...omentum-2.html

You have to follow the same suggestions as before.