CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Eddy Dissipation Model (

Fred June 28, 2006 03:27

Eddy Dissipation Model
The eddy dissipation model coefficients A and B can be entered globally for the whole domain and seperately for each reaction. Which one does the solver take or do they interact with eachother [A=2(globally) and A=1.5(reaction) => A=3(overall]) ?

opaque June 28, 2006 08:17

Re: Eddy Dissipation Model
Dear Fred,

The hierarchy is based that global coefficients are used only when the coefficients are not defined at the reaction level. There is no other interaction.

In your case, the CFX solver will use A=1.5..


PS. Similar hierarchy is for local coordinate frames.

Fred June 28, 2006 12:49

Re: Eddy Dissipation Model
Thank you Opaque.

I hope you can help me with an additional question. How many reactions can CFX(EDM and Finite Chemistry) handle? I'm using 4 reactions including backward reaction rates but i'm getting domain imbalances about 100% for some materials.

opaque June 28, 2006 12:58

Re: Eddy Dissipation Model
Dear Fred,

There is no limit (besides memory or a bug) for the number of reactions in the CFX solver..

Mixing EDM with Finite Rate Chemistry reversible reactions is a tricky bussines. Using the EDM model for more than 1 reaction is already a "gray area", adding reversible reactions is even darker..

The CFX solver does not account for an EDM backward reaction rate.

I am not sure what you are trying to do..


Fred June 28, 2006 14:42

Re: Eddy Dissipation Model
Hello Opaque,

I'm doing a "simple" combustion. I want to predict better dissoziation effects. Using only forward reaction rates they are too much depending on temperature. Would you suggest to use only forward reaction rates and slow down reaction speed?


Marta May 4, 2010 11:08

I'm simulating a hybrid rocket combustion process in CFX, and apparently the eddy simulation approach would work in an appropriate way.

But i have some problems with temperature, because the flame temperature exceeds the adiabatic flame temperature limit. I know that this can happen in CFD simulation, anyway these results are unacceptable for my purpose.

Therefore I tried simulating the chemistry with a different reaction, which is less energetic than the first one i applied and also more reliable, according to the species mass fractions i had using CPROPEP as a control means.

At this point, what happens is that the flame creeps on the combustion chamber walls, but i saw that this can be a drawback of the eddy dissipation model in some cases, because k tends to zero at the walls and thus the epsilon/k parameter can become singular.

Then i tried to solve my trouble using the 'limiting mixing rate' as it is presented for methane/air flames, because i don't know how to set it correctly for paraffine. But the simulation converges with some difficulties and not with as low residuals as without activating this parameter.

Has anyone dealt with this kind of problems? Can anyone suggest me something else to try?

Thanks in advance!


All times are GMT -4. The time now is 14:54.