[Whitebear]

During coal combustion steady state analysis, I meet convergence problems in heat transfer and mass fraction. Momentum and mass, turbulence, and particle source change rates satisfy default residuals. And monitoring point also shows steady state characteristic well. How can I improve convergence in heat transfer and mass fractions? I tried timescal change in mass fraction as autotime scale and physical timescale but it didn't help. Fuel Gas and H2O RMS Mass Fraction especially show bad residual plot 1.0E-3<residual<1.0E-2 and almost constant lines. Thank you for reading my problem.

[ghorrocks]

You have to expect convergence difficulties in combustion modelling, especially for the heat equation and/or combustion reactants.

First of all, have a look at the residuals in the post processor. You might need to add the residuals to the results file to do this. Have a look at where the maximum residuals occur - is it at an area of importance to the flow?

Also consider whether your convergence is tight enough, even though it has not reached your default convergence tolerance. Do a sensitivity study and see if you need tight convergence or not.

[Whitebear]

Thank you very much indeed ghorrocks. Due to your help my work has developed quite a lot. I want to ask another thing related to coal particle emissivity setting in CFX. I found some papers dealing with particle emissivity. It changes with furnace temperature. So, I put it as temperature function but CFX says 'Only CONSTANT emissivity is supported for particle materials.' Do you have any advice for helping me again? As you know coal particle emissivity affects a lot of combustion calculation and residuals.

[ghorrocks]

I did not know that actually. I don't work in the field of coal combustion - but I can see how it has an effect.

First of all I would do a quick sensitivity test - run a high and low emissivity and see if it does make a difference. If it makes no difference then it does not matter. You could also try setting the emissivity at the average temperature of the furnace, that might be close enough.

There are ways around the emissivity as a function of stuff restriction but better check the basic stuff first.

[Whitebear]

As you said, I am trying to do the sensitivity test. As I told you, emissivity takes a lot of effect of convergence and calculation results in coal combustion problem. Thank you very much for your help. It is almost impossible to solve the problem properly without proper boundary conditions like coal combustion tutorial.