[Falcon16]

Hello,
I am trying to run a simulation of a axial turbine stage with a fourier transformation. The pitch is very high (~2.5) so its not possible to use time transformation.

When i start the simulation run from a steady state initilization file the first 2000 timesteps (Initillization Phase) pass without any problem. But as soon as the startup phase beginns the simulation starts to become unstable. The Mach Number increases while the residuals remains acceptable (RMS ~1e-4). In addition the Courant Number starts to rise. After about 20 Timesteps in the Startup Phase the simulation crashs.

One time a simulation survived the startup phase but then crashed in the Full Fourier Transformation Phase in the same way as described above.


The Courant Number of the simulation is ~2. Thats not perfect but for computational power reasons there is no choice. Pure Transient simulations of similar meshes run with the same Courant Number without problems. The mesh quality must be good. The mesh worked with pure transient simulations without problems.

I tried nearly everything to solve the problem. I used the Damping Factor (1.0) and incresed the Fourier Coefficients to 10.


My setup is a transient rotor stator case similar to the one in the tutorial(witch by the way worked without problems) but with much higher pitch and higher pressure. In addition the mesh is about 50 times more detailled.

There must be something wrong with the collected Fourier Coefficients because the simulation crashes as soon as they are used.

I am grateful for any help.