[xxzr_CFX]

Hello everyone,
My research object is a singel stage compressor with two phase inlet condition(99% Air perfect gas and 1% water droplet), and I have got many steady results. But when I try a transient calculation with steady results as initialization, it always arises a maximum mach number warning after about 300 physical timesteps, that is:

| Notice: The maximum Mach number is 2.522E+00. |

After several timesteps of the above warning, there will arise a Newton's method warning, that is:



The calculation will crash right away after the Newton method warning:



My basic setting cotains: k-epsilon turbulence model, high resolution for both advection scheme and turbulence numerics, fine mesh, 30 timesteps for one blade period, 20 coefficient loops for one timestep.

I have changed several expert parameters, but it did not work, which contains:
1. Newton Pressure Under Relaxation of the mixture: 0.02;
2. model coefficient relaxation in expert parameters: 0.2

Looking forward to your advice!

[ghorrocks]

If it converges steady but fails transient then your time step size is probably too big. How did you choose your time step size? I recommend you use adaptive time stepping with a minimum time step size set really small (1e-10s or whatever is appropriate), and homing in on 3-5 coeff loops per iteration and let CFX find the time step size required by itself. Also use a really small initial time step size and let it increase to the target amount rather than be too big because it might crash before it gets to an appropriate size.

Also - put all those expert parameters back to default. You should not need them. Also also - save a results file before it crashes and have a look at how it is progressing. It might give you some clues as to what area in the domain is causing the problem.

[xxzr_CFX]

Thanks for your advice! I will try it right away.