I am trying to model a 10-step reaction where one of the reactions is H + O2 --> OH + O

This reaction leads to unrealistic species concentrations every time I turn it on. What can I do to improve convergence?




CFX10 has a stiff chemistry solver that should be used every time reactions with widely varying timescales are being solved. This solver linearization the chemical source terms and uses partial coupling between temperature and species (by assuming a constant total enthalpy perturbation) to keep reaction rates under control. False timestepping allows the shorter reaction timescales to be solved with the same timestep as the hydrodynamics without decoupling. The stiff chemistry solver is controlled through the use of EXPERT PARAMETERS, as described below:

**********************************************************************
All cpp control parameter macros begin "__stiff_chemistry_".
Values in square brackets are the default settings
**********************************************************************

stiff chemistry : logical [F] used to switch on stiff chemistry code.
stiff chemistry full Jacobian : logical [F] used to turn on a full sensitivity Jacobian.
stiff chemistry cutoff mf : real [1.0e-6] used to provide a cut off mass
fraction in rate and other calculations to
ensure physical behavior as concentrations go
to zero. Because the user's physical model is
not respected below this value it should be
smaller than any value expected to be of
physical importance in the problem being
solved.
stiff chemistry fts urf : real [0.9] used to under-relax the false time step contribution to the diagonal.
stiff chemistry expt urf : real [0.5] used to 'under-relax' the explicit
time step limiter for the false time step.
This limiter ensures that the false time step
is not smaller than would be obtained by
using an explicit time step divided by this urf.
stiff chemistry sol change fac : real [0.6] used to control the maximum solution
change allowed in mass fractions and energy.
The maximum is computed from the coupled mass
fraction system and no mass fraction Y is
allowed to change by more than fac*Y if Y>Yc
where Yc is the cutoff mf.





Show Form
No comments yet. Be the first to add a comment!