Does anyone know which reactor type Fluent uses in the detailed reaction modeling when using the Eddy Dissipation Concept (EDC). In the manual it only states: "In FLUENT, combustion at the fine scales is assumed to occur as a constant pressure reactor, with initial conditions taken as the current species and temperature in the cell. Reactions proceed over the time scale (tau*), governed by the Arrhenius rates of Equation 14.1-8, and are integrated numerically using the ISAT algorithm [290]."

But a constant pressure reactor can be both PSR or plug flow. The theory behind the EDC model states that PSR reactors should be used, but it would probably be a lot easier to reach convergence with a PFR type reactor.

I would like to compare different combustion mechanisms, but which reactor type should I compar them in?

Jimmy,

Isn't the PSR (I assume you mean Perfectly Stirred Reactor) just a set of assumptions for i.e. calculation of combustion efficiency.

No offense but I can't see why you need to know this. If you want a serious turbulent prediction then use the pdf transport model for a detailed mechanism - EDC is good but has been around for decades.

The flamelet model requires a detailed mechanism but you need to decide whether your model should be premixed, non-premixed, or partially premixed.

The EDC has the advantage that it considers briefly what effect turbulence has on the flame, but again the pdf transport method is far superior to the EDC but also ridiculously expensive. In fact around 1 hour per iteration on a newish laptop (1Gb memory) for 100,000 cells with a 37 species mechanism!

The flamelet model gives optimum economy.

Phil

Hi Phil

Thank you for your answer. You are right that the PDF transport model is superior, but the computational expense makes it useless for practical applications (for now).

The flamelet model has become a lot more interesting with the new unsteady post processing feature in Fluent 6.3, but still it assumes fast chemistry for the steady modelling part, which can be problematic for low temperature CO burnout (which is part of what I am modeling).

Besides I like the concept of the EDC model, which decouples the chemistry part and treat it on a fundamental level.

/Jimmy

You're using an eddy dissipation type model for CO burnout? I'll eat my hat is you even get reasonable results. It WILL over predict the temperature.

Most people use a cluster to use the pdf transport model. How can something be more practical when it gives inaccurate results?

The NOx model can be used to predict slow burning flames.

Phil

The EDC model in Fluent uses a plug flow reactor

[hashim84]

I am running some experimental cases with EDC, and if I run the same cases with PDF(Non-premixed Combustion) model in Ansys Fluent, the adiabatic flame temperature would be different for PDF models (1800K), but in EDC ( T= 2200K and above). So the difference is around 400K or more sometimes. Just for your information ( everything which i am using is the same like the geometry, meshing, and the boundary conditions, etc) Only the model is different. So, i need your experience guys to figure out why this difference in temperature.
Also, I want to run one case in PRO II software, and see what would be the temperature, what kind of reactor should i use?
Thank you so much in advance!

[chittipo]

EDC model has two key model parameters which should be noted here. The volume fraction constant and time scale constant. Using the default values could be a reason for the over prediction of the temperature. I think these model parameters should not be treated as global constant. For e.g., increasing the time scale constant from the default value (I think 0.408) to several orders in magnitude could bring down the temperature. As it delays the reaction. I think what are the correct model parameters for a given system is the difficult question.