Author Topic: Cavitation model with dynamic mesh doesn't converge!  (Read 993 times)

Offline pitney1

  • Jr. Member
  • **
  • Posts: 62
  • Reputation: +0/-0
  • Searching for solution
    • View Profile
Cavitation model with dynamic mesh doesn't converge!
« on: April 10, 2012, 10:27:45 AM »
Advertisement
Hello,

Cavitation model with dynamic mesh doesn't converge!!! The case carries pressure bcs.

Please help.


Offline william

  • Full Member
  • ***
  • Posts: 151
  • Reputation: +14/-0
  • Know it, share it.
    • View Profile
Re: Cavitation model with dynamic mesh doesn't converge!
« Reply #1 on: April 10, 2012, 10:29:20 AM »
It is often found that cavitation model is not converging. This is also found true when dynamic meshing is also used with it. Thus it becomes a transient simulation. In this type of cases, the main problem is that when the cavitation model is on, there is a considerable reverse flow at the inlet. By default, the cavitation model does not allow reversed flow at the inlet to speed up convergence for many cases. But here, such a condition proves to be too strict and too difficult to satisfy. To accommodate this, a remedy is given in the code (FLUENT-6.2). The modifications in the solver are focused on the pressure BCs in the cavitation model. In the cavitation model, the basic formulation does not allow reversed flow at any "PRESSURE INLET". It was found such a practice applies for most cases and make the solution process more stable. One can allow for the reverse flow at inlet even with the cavitation model by using the following scheme/text command on FLUENT's console window.

(rpsetvar 'cavitation/pressure-bc? #f)

by default the above value is TRUE. (i.e. #t)

To check this , type

(rpgetvar 'cavitation/pressure-bc?)

This will return either #f or #t.

The above rpsetvar is available in FLUENT-6.2.3 and later versions.