How well can the CFX 10.0 cavitation model account for thermal suppresion as the fluid temperature increases?



It is true that the thermal effect in the cavitation modeling in CFX-10.0 are not taken into account. In CFX-11 the model does take into account the thermal and turbulence effect. It is a beta feature and it can be set via the GUI (by enabling beta feature under cavitation model). for the thermal effect, there two models: the full thermal model and the Taylor model.
The main important steps are:
- The user has to set the enthalpy for each phase
- The user has to set the pressure of saturation as a function of the temperature. This can be done via CEL. In the example, the Psat is function of T using the Antoine equation.

The general guideline for a cavitation simulation (to calculate the drop curve) is the previous res file should be used as initial guess for the actual run:
For example

Run#1: atdesign point and no cavitation with the following BC's: Ptotal at the inlet and Mass flow at the outlet

Run#2: at design point and with the cavitation model on using the res file obtained from run#1 as initial guess and dropping the Ptotal at the inlet by certain amount (for example 0.1 bar)

Run#3: At off design point with cavitation model on using the res file obtained from Run#2. If this is far (70% of design for example or less), the user has to drop the total pressure at the inlet with a small step (I have to reduce the total pressure by a step of 10000 [Pa] or even less near the drop point NPSH%). Since he knows the experimental NPSH%3, the user can starting reducing the pressure by a small step around the NPSH%3.

Also, the user may consider increasing the RMS to 1 magnitude higher because the strong non linearity of the cavitation model at off design point which may causes a problem of stability of the solution.





Show Form
No comments yet. Be the first to add a comment!