Learning Forum

[Kauê Castro]

I am trying to simulate a periodic excitation on a jet but Im having trouble making two boundaries work together, only one works when I input them both. For the excitation Im using a "Named Expression" as follows on the image and for the other, Im using a developed internal flow I've previously simulated, as follows:

As you see, Im using a step function for the periodic excitation, while the value is 0 for the excitation, the jet occurs normally, however, when the 2 seconds threshold finishes, the jet plummets, I have recorded the animation but is not possible to post here, therefore Im posting a before and after frame literally:

4I'm running turbulent & transient, with a time step of 0.01 s.

[NickFL]

The sin goes negative right? so basically your jet would be going in the other direction. Your inlet is now an outlet. And what I imagine you are seeing is a divergence. Look at the scale of the last image. You have velocity of 2500 m/s. The jet is probably still there, but the scale (and the divergence) is ruining the picture (and calculation).

[Kauê Castro]

I see, but the problem is that, using a smaller value for the pressure deviation, it works fines and converges (although it's effects are very small).. I'm having a hardtime troubleshooting when the simulation doesn't converge, I fiddle with the initial values and the turbulence parameters which in actual fact is almost always the reason of divergence, it's very common I get a "Turbulence Viscosity Ratio limited to 50000" message while running it. Im thinking now about possibly the Courant value is a bit high, or maybe the relaxation factors are not "optimal"... If you have any suggestion, it would be greatly appreciated!!
Thanks for the response NickFL :)
Sincerely
Kaue

[NickFL]

I think what you are experiencing is divergence. When you start having the "Turbulence viscosity ratio" warning, it typically means things are going awry. How does the mesh look? When moving to a higher amplitude of pressure oscillations, have you also reduced the time step? The root cause could be several things, so you will have to do some debugging.