[dancfd]

Hello all,

I am running a 2D airfoil simulation with a C-grid which was going fine until I tried to change in magnitude of the freestream speed as a funtion of time at the inlet. Now, the simulation crashes with the following error:

[3] --> FOAM FATAL ERROR: 
[3] Maximum number of iterations exceeded
[3] 
[3]     From function specieThermo<Thermo>::T(scalar f, scalar T0, scalar (specieThermo<Thermo>::*F)(const scalar) const, scalar (specieThermo<Thermo>::*dFdT)(const scalar) const) const
[3]     in file /opt/openfoam211/src/thermophysicalModels/specie/lnInclude/specieThermoI.H at line 69.
[3] 
FOAM parallel run aborting

Others have had this error, and so I checked some of the other recommended fixes on the boards:
- increased specieThermo.C "const int Foam::specieThermo<Thermo>::maxIter_" to 5000 from 100 (no effect)
- Changes relaxation factor on e from 1 to 0.3 (no effect)
- Reduced Courant number to 0.3 from 0.5 (no effect)
- Checked number of iterations and residuals on e (no change from previous timesteps)
- Changing the time discretization scheme from backward to Euler (delays the problem to a later timestep but it still prevents the case from completing)

I plotted the two attached graphs to identify deviant cells. It appears that a small number of cells are diverging.

Other info:
- OpenFOAM 2.1.1
- solver: rhoCentralFoam
- ThermoType: ePsiThermo<pureMixture<constTransport<specieThermo <eConstThermo<perfectGas>>>>>

Not sure how to change the case to make it converge. Thanks in advance for any suggestions anyone can offer!

Regards,
Dan

[masoudsh]

hi Daniel

i have your problem too.
did you solve it

regards,
masoud

[olivierG]

hello,

If reducing time step doesn't help, and since you use perfect gaz model, check min/max of pressure: you may have some cell with bad/crasy value.

In this case, a rude workaround is to limit pressure like p = min(p,1) i.e 1 Pa min here
You need to modify the solver code for this.

regards,
olivier

[masoudsh]

hi Olivier

thanks for reply to my question.
i reduce my time step and it work but it .000001 and very low.
the checkMesh is ok.
you said "you need to modify the solver code for this",how can i do that?

regards
masoud