Hi, I am trying to model a large-scale non-premixed jet fire in FLUENT, I am using the laminar flamelet model. I am trying to use LES. The model runs fine with RANS. If I use my converged RANS as initial conditions for my LES within a few timesteps the flame dissapears. I have no idea what I am doing wrong. If I look at my fuel concentration at the jet inlet it drops from a mass fraction of 1 (RANS) to less than (0.4) LES in a matter of time steps. There seems to be nothing burning.

I've been trying to get to the bottom of the models in Fluent to see what changes with LES with limited sucess.

Please help!!

Not an expert in Fluent: however you can not couple RANS combustion models straightforward to LES.

Have a look a their manuals and look how they do it in LES case. After that you will see whats going wrong.

Thank you for your response. I'm working through the manual again incase there is something I missed (quite likely!). However, I used RANS as an initial condition for LES on Fluent's suggestion. I was originally starting straight from LES and was seeing problems at my inlet boundary, the mixture fraction was changing from 1 to 0 within the first few cells. My fuel seems to burn instantly. I observe this same behaviour when I move from RANS to LES and is why the flame "dissapears". The PDF lookup table is calculated in advance and does not seem to be the problem as it is the same for RANS and LES.

Although it is calculated differently in LES and RANS the mixture fraction variance seems to be similar in the vicinity of the inlet boundary in each case. The difference is in the mean mixture fraction.

There seems to be something wrong with my LES model set-up which is affecting my mixture fraction calculation. I am sure it is something obvious I just don't know what!

for les.....the initial condition should have some random fluctuation on mean profile u got from RANS....try...

Thank you very much for your suggestion. However, I have been applying a fluctuation at my LES boundary, I've varied it from 0 up to 30% with no success unfortunately.