Hej, I want to model combustion. I've set up the case with the eddy-dissipation-model. Turbulence modeling is covered by rke and enhanced wall-treatment. y_plus is around 1. The flow field converges (including the dpm-data). But the moment I enable the energy equation to be solved, the residuals for k and e become very large. Fluent detects divergence. Trying to lower the relaxation factors for turbulent quantities makes Fluent diverge in respect to temperature or causes a NAN-Error (it depends...) The temperature exceed the 5000K-limit as well. Has anyone experienced this before and could give me a hint, please? Regards! Joerg

hi, you have to verify matierials properties .Can you specify what type of fuel you use? If you are using methane you can try to put the dilute approximation to constant.

I use liquid diesel (c10h22) as droplets. The dilute approx is set to be constant. The fuel mixture consists of gaseous diesel phase (fuel), o2 (oxidizer), n2. The reactants are co2 and h20. The stochiometric values of the reaction are specified correctly. Thanks for your input... I'll double-check the materials!

Hello all of you, I've managed to fix the divergence problem. But now there is no evaporation of the liquid phase. Combustion takes not place, therefore. Any hints would be very appreciated! Regards, joerg.

try patch.........i.e add a energy source to your mesh........this will not affect your final solution........ hope it helps........

I've patched the injection zone to 2000K already. Nothing happens... My next try was raising the temperature of the flow at the inlet a small value (4 Kelvin) above evaporation temperature. This doesn't help either. Hmm... I'm clueless... Any other help? Thank you! btw: Is it normal that the patched volume "disappears" after patching a certain value into this zone? Thanks for sharing your ideas, Hitesh! Regards, joerg.

[pghoseju]

OK I will give you a suggestion. Surely you will get evaporation. Go to material and increase the value of binary diffusion of diesel droplet liquid from 3.66e-6 to 3.66e-4. then watch

[kingpots]

Are you using unsteady tracking and injection ? if yes try to increase the size of the time step
also the convection/diff. option will affect the evaporation (do not use it ) use only diffusion