Learning Forum

[Saeideh.Kermani-Afarani]

Hello Dears,

I am simulating the oil supply of a wet-running multi-disk clutch in ANSYS Fluent. The model is divided into two domains: the stator (outer disks that do not rotate) and the rotor (inner disks that rotate). The rotation axis is aligned with the X-axis. I have defined symmetry conditions for the model and performed the simulation.

I expected the total pressure to increase in the radial direction due to centrifugal force. However, the results show the following: The velocity vectors are directed tangentially. The fluid flows into the lower symmetry plane and exits through the upper symmetry plane (flow in the tangential direction), which is as expected. However, the static pressure does not increase radially but rather in the tangential direction. This could be causing fluid buildup behind the upper symmetry plane, where the fluid should exit. The total pressure distribution resembles the static pressure pattern. However, the dynamic pressure in the rotating domain seems to be directed radially and increases in the radial direction.

Interestingly, when I simulate the entire model (without symmetry conditions), the pressure contours match my expectations.

Additional Information:

• Gravity is turned off in the simulation.
• The simulation is a steady-state simulation with frame motion.
• Inlet type: Mass Flow Rate
• Outlet type: Pressure

Question: Why are the symmetry conditions being ignored in this case?

Best regards

[Rob]

How can you have symmetry at the inlet/outlet part of the system? Rotational periodic may be a better option. 

[Saeideh.Kermani-Afarani]

Thank you very much for your response.
The full-circle model has 16 inlet pipes, while the symmetry model has 8 inlet pipes (all of them are grouped into one inlet).For the inlet type with mass flow rate, the mass flow rate for the symmetry model is halved and entered accordingly. The full-circle model has 6 axial outlet pipes (all of them are grouped into one axial-outlet) and 6 radial outlet pipes (all of them are grouped into one radial-outlet), while the symmetry model has half of that: 3 radial and 3 axial outlet pipes.For the outlet type with pressure, the ambient pressure is entered (for both the symmetry model and the full-circle model).

Is this what you meant?

[Rob]

Yes, but symmetry is a boundary that reflects everything, how is a rotating domain symmetrical? 

[Saeideh.Kermani-Afarani]

Thank you for the response. The symmetry plane consists of two parts. One part is in the rotating domain, and the other part is in the stationary domain. Both symmetry surfaces are connected together but are defined separately as two symmetry conditions.

[Rob]

If the domain is rotating those surfaces are not symmetrical for the flow: have a look at rotational periodic boundary type.