[Manuelo]

Hi everyone,

I'm performing a combustion calculation of a boiler using Fluent. There are 6 swirl H2/air burners. I get reasonable results with SST steady calculation. A brief setup of the case:
- SST for turbulence.
- Around 3 millions cells, poly-hexcore.
- DO with participating media (H2O).
- Ideal gas for density (fuel nozzle is at very high velocity), mixing laws for cp, viscosity and thermal conductivity.
- EDC with a mechanism of 10 species and 21 reactions. Direct integration.
- 2nd for all but DO with 1st order.
- Simple for press-vel coupling.
- Least squares cell based.

This setup reaches a good convergence condition. In particular, heat imbalance is 3 orders of magnitude lower than heat flows to the walls or through inlets/outlets. Rest of the model results make sense to me.

When I switch to transient calculation with SAS for turbulence, I get results that while initially look rather reasonable, once I inspect it further, there is a huge heat imbalance of 30 MW, which is around 15% of the main heat flows in my problem. Gas temperature are lower than in the SST steady case. Heat to the walls decrease as well. With respect to the setup of the SST steady case, I have changed the following:
- SAS for turbulence.
- PISO for press-vel coupling
- BCD for momentum.
- Bounded second order for time integration.
- 0,001s of time step. I have tried lowe values also with the same problem
- 1 DO iteration per each 5 flow iterations, which ensures DO residual to go below 1e-6.

I have gone through the setup modifyng several things with no success to get the heat balanced:
- Coupled for press-vel coupling.
- time steps as low as 0,0001s.
- More DO iterations to reach energy and DO residuals down to 1e-7.

I tend to thing that I am missing something that has to do with using SAS with EDC in the same calculation. I'm currently trying transient with SST and SAS with ED, just in case I find any clue about what I am doing wrong.

I would appreciate if anyone can give advice/hint. Thanks!!