Learning Forum

[kyshikawa]

Hello everyone,

Ansys beginner here. I'm practicing with heat transfer, and I have a problem I can't seem to do right. I have a solid cylinder (rotor) inside of a cylindrical shell (stator) which will be cooled by water (fluid) that flows axially inside the "water jacket". The rotor will be generating 30W of heat (8153.3 W/m^3) and the stator will be generating 350 W (100981 W/m^3). The solid parts are made of steel with K=16 W/m.K. Water will be flowing in at a rate of 0.06615 kg/s, entering an leaving at atmospheric pressure. The convection coefficient between the stator outer surface and the fluid is 115 W/m^2.K. As a condition, I want that the surfaces (that are in contact) of rotor and stator are at 80 degC. 

Up to the meshing, everything is fine. However, I open fluent, I see a lot of walls I know, and that when I click on, don't even show because they don't exist. I also see a list of contact regions with walls associated. Now I don't know to which wall exactly I should assign the temperature values. I tried assigning the values to the walls related to the contact surfaces, but when I run it I get the maxim temperature around 60 degC. (Which cannot be, since I set 80 degC). 

Legend: pink is the jacket, green is the fluid, red is the stator, and orange is the rotor.

I would really appreciate help understanding this walls that Ansys automatically creates. Perhaps I'm assigning the values wrongly. Thanks a lot in advance!

[peteroznewman]

I have also been learning FLUENT and my advice is to open the Geometry editor (SpaceClaim or DesignModeler), and use Shared Topology. In SpaceClaim, that is done by using the Workbench tab and clicking on Share. The two faces on each side of the water body should light up. In DM, you select the three solid bodies in the Outline and Form New Part.

Next, drop a new FLUENT analysis system on the Geometry cell. After it copies the geometry, you can delete the link between the systems. Open the Mesh cell and create the named selections for the faces like inlet and outlet. Make sure there is no contact defined in the Connections folder. If there is, delete it. Make sure the three bodies are properly identified as Fluid or Solid in the Details window.

Now when you open FLUENT, you should have a simpler list of Boundary Conditions.  The wall that is between the water and the inner cylinder will have a shadow. That is so you can define different properties for the water and the solid. Same goes for the wall between the water and the outer cylinder.

 

[kyshikawa]

Hi Peter,

Thank you very much for your post. I managed to remove the extra walls, and the solution I get is vey close to what I calculated. 

Now I have encountered a new problem. I wanted to see the fluid streamlines, and see how the inlet temperature varies as it approaches the outlet. However, when I create the streamlines plot I get the following:

Then I tried to do a vector plot, and this is what I got:

I don't know what can be causing this weird flow (or not flow). Also, here's a screenshot of my new wall list. I noticed that my fluid walls are together in "wall-fluid". Perhaps this is the issue? I only put the convection coefficient on the inner cylinder surface next to the fluid. Should I have also done that with the fluid walls?

[kyshikawa]

Hi Peter, 

can you please help me the streamline plot? Thanks a lot in advance!

[Rob]

You only add a convection coefficient etc on walls that connect to outside of the domain: it determines how energy can enter the model.  Please plot a velocity contour through the length of the model (use an isosurface) so we can see whether you have flow. 

[kyshikawa]

Hi rwoolhou,

here's the isososurface

So it seems my fluid is stationary (though I gave an inlet mass flow), and also by some reason, the fluid is discontinuous.

[Rob]

No, a velocity contour on a plane (isosurface of y=0 for example), I'm trying to figure out what's meshed.  But, yes that result looks very odd. 

[kyshikawa]

Hi rwoolhou,

sorry for the very late reply. I managed to do the contour on a plane along the shaft (I hope this is what you were referring to). It seems my fluid does come to a complete stop, or disappears, and then reappears at the end of the motor.