[zhyltx]

Hi All, I am trying to implemented pressure-dependent reactions into CFX? But the results from CFX do not match those from CHEMKIN... Has any one successfully done that?

From the Chemkin manual, the equations are not difficult to understand. I follow the Chemkin manual to build-up expressions in CFX, and then set forward and reverse reaction rate with corresponding expressions. The calculation is based on Chemkin manual. Also can be seen in the following:

http://www.ems.psu.edu/~radovic/ChemKin_Lindemann1.pdf

The temperature rise given by CFX is lower than that given be Chemkin. I guess something might be wrong with reaction set-up in GUI. Is there special something that needs to take care?

I really appreciate any comments or suggestions. Thank you very much in advance! We can also contact via email.

Jerry

[ghorrocks]

Have you checked the basic stuff? http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

[zhyltx]

Originally Posted by ghorrocks

Have you checked the basic stuff? http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

Hi Glenn,

Thanks for replying. I forgot to mention that what I am testing is a single cell domain with a reaction, in other words, non-dimensional test. All boundaries are adiabatic walls. It is laminar, no velocity and with time-step of 1.0e-08. High order scheme is used and convergence is very good. So, I guess the problems mentioned in that website are not the issues in my case.

I think there is something wrong with using expressions to give values for reaction rates. I also found that if I use GUI interface to set a reaction (finite-rate chemistry model), the results are good and match with Chemkin for single-step reaction test. But if using expressions to give values for forward / reverse reaction rate, the results are different with that using GUI...

So, I would like to ask what is your opinion about that? Any comments would be appreciated.

And has any one successfully used expressions to implement Chemkin into CFX (single-step reaction or more complex one)? Please share some experience if possible. Thanks very much!

Last, do you know are there short names of NASA polynomial for species, like O2.a2 or something? Inputting them manually into expressions is, I feel, a big job and easy to make mistakes...

Best Regards,

Jerry

[ghorrocks]

I have no experience with chemistry modelling (with CFX or anything else) so cannot comment on that.

The link I suggested also has comments on mesh, time step and convergence criteria so you need to check they are OK before concluding anything.

It is quite likely that the equations are coupled to the flow differently from CEL to the built-in models. Hopefully this would result in a difference in convergence but not a difference in the final result.