Hi,

thank you in advance for taking your time and energy to give me your valuable proposals.

The problem is as follows.

I am at present doing a non premixed gaseous combustion air/methane using Fluent 6.1 on a 2D problem in order to validate some experimental data obtained ealier.

I used a PDF modelling approach but the results are clearly underpredicted by around 200-300K. I tried to change the BC at the walls and inlets in term of temperature, mean mixture fraction and mean fraction variance but it only accelerates the convergence without affecting the results.

I used a steady state simulation and previously carefully defined the parameters using prePDF which has then been implemented into Fluent. The prePDF results seem correct. In Fluent I activated the DTRM radiation model.

The computations seems to be going well and obtain convergence after rougly 4-500 iterations. I also refined the mesh without better results so I cannot validate my model.

Could someone please provide any suggestion on what might be wrong in my problem settings.

Any suggestions are highly appreciated.

Thank you very much

David

Don't forget to use the Weighted Sum of Gray Gases for the calculation of the gas absorption coefficient.

I use the D-O method for radiation, but DTRM should work well too. Radiation calculations take a long time to converge (my experience), perhaps 500 iterations is not enough.

thank you very much for your insight and the time taken to answer my problem.

Indeed the radiation calculation take more time to converge than i previously thought and I didnt take into consideration the gas absorption coefficient, which would probably significantly inflence my results.

the informations you've given me are very useful

David