[javarocket]

Hi everyone!

I am trying to carry out a transient flutter analysis for the rotor of a subsonic axial compressor with 17 blades, so I first did the steady solution in order to initialize the transient one. I, of course, modelled 2 passages, instead of the whole rotor using periodic and sampling interfaces. Later I will refer to this analysis ad "only rotor".

The steady state analysis run without any problem, instead the transient one, using the Fourier Transform, run but I don't obtain proper values for the Aerodynamic damping and a fatal overflow occurs after roughly 1500 timesteps.

Using the Time Transform didn't occur a fatal overflow, neverthless the analysis did not converge and the aerodynamic damping values were not reliable.

In the end, I tried to model the whole rotor (17 blades) without sampling and periodic interfaces, without Transient Blade Row (TBR) model (just a "transient analysis") and defining the aerodynamic damping monitor manually.
Using this setup I finally obtained a good result for the aerodynamic damping and the analysis seems to be converged.

Does anyone have any idea why the Fourier Analysis is not working properly?
Non dovrebbe essere il migliore per eseguire l'analisi del flutter?

Later, I tried to simulate a full stage analysis (rotor + stator downstream), with 17 blades (not just 2 passages), with mesh deformation and transient rotor-stator interface.
I used TBR model and Fourier Transform, using the mixing-plane between rotor and stator as a domain interface for the FT.
With this setup I obtained reasonable values for the aerodynamic damping, but the problem is that I do need to carry out the "only rotor" analysis with the same setup in order to compare the results and to decide if they are correct or not.

Is anyone able to explain to me what is happening?
Thanks for your time!

javarocket