[jose_rodrig]

Hi,

Im trying to simulate combustion inside a CAN Type Gas Turbine Combustion Chamber with methane (CH4).

In case you know, the input conditions and the mesh are similar to the 3-step EDU tutorial from the straccm+ manual. If thats not the case, shortly, the setup consists of a 60 degree wedge with 2 swirling air inlets to the main chamber, 1 swirling fuel(CH4) inlet and one 3rd air inlet running in an outer channel for cooling. Both the 2 air inlets and the fuel inlet are characterized with cylindricalInletVelocity BC, supplied with OF-1.7.1.

MESH
The Can is 1m long and .25m radius and the domain only represents a 60 degree wedge section of the Can. The mesh as 1800 cells (coarse mesh). The wedge sides have symmetryPlane BC.


I tried to run this simulation with reactingFoam solver modified to account ignition (as purposed in the OFwiki).
Plus, most of the settings of the several dicts were taken from the counterFlowFlame2D tutorial (such as initialChemicalTimeStep and other coeff). RANS was used for turbulence modeling.


RESULTS
At 1st, I used the ode chemistrySolver in the chemistryProperties dict but i got the janaf out of range error. The temp exceeded the 5000k limit.

Next, I used the sequential chemistySolver and then I got a solution. However, the residuals were very unstable with a courant no. of 0.5 and the MAXtemp would oscillate between 2600k and 3100k.

QUESTIONS:
1. What are the basis of either the sequential and the ode chemistrySolvers? Can you give some references to this models?

2. Which Courant number should I set?

3. Can I use different ddt schemes such as CrankNicholson with reactingFoam? What about steadyState?

4. Which settings would you recommend to tune so I can achieve convergence. I m avoiding refining the mesh to keep the setup as close as possible to the tut in starccm+.

5. Is it possible to use dieselFoam without injectors or spray?

That is all =P
I sincerely hope this thread will be a contribution to the OFOAM community!
Thank you.