PDF - conserved scalar approach in KIVA

vega

Dear all,

I am trying to simulate turbulent non-premixed flame with equilibrium chemistry. So I`m using pdf-conserved scalar approach (equilibrium flamelet).

I have implemented PDF-mixture fraction in KIVA code and obtained all reactive scalars (temperature, enthalpy, density, and species concentration) as functions of the mixture fraction.

I already compared the density (obtained using pdf) with the density obtained using cold-flow KIVA simulation for several time steps, and they are in excellent agreement.

As far as I understand, the density must be returned back to KIVA to calculate the turbulent flow field, so I can obtain new values of k and epsilon to be used in the transport equation of variance of mixture fraction which is required to construct pdf at the next time step.

The current problem is, after I coupled the new density with KIVA (say.. if ro(i4) is density of original kiva, and rbar(i4) is density obtained using pdf, I simply wrote ro(i4)=rbar(i4) ), the code crashed due to temperature overflow.

I am trying to trace back the error in the code list, but I was overwhelmed with the complexity of the code.

So the questions are: 1. How can I couple the density (obtained using pdf) with kiva?

2. Since original KIVA also calculates the species concentration, internal energy, and temperature, but I already obtained them using pdf, do I have to deactivate all of the KIVA subroutines associated with these scalars?

Thank you for your kind help.

regards,

vega