CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   We have to choose a CFD (

pierrel September 24, 2012 02:58

We have to choose a CFD
I'm involved in a "Eco Marathon" Project (or supermileage...)
The project consist in building a prototype that has to be as effective as possible.
This is the site of our project:

We would like to process the body of our prototype in a CFD package in order to try to find improvements.

Do you have a CFD software to advise us ?
Price, ease of use are of course importants criterias.

To be complete, the body has been designed using Autodesk Inventor .

Please do not hesitate to sk questions if my question is not accurate enough...

Thank you in advance,


Maged_CFD September 25, 2012 00:33

Hello Pierre,

I checked your website and it seems like a very interesting project.

If you are looking for an accurate, easy-to-use, fast and cost effective CFD package to simulate your prototype, please check out SimericsMP CFD software.

SimericsMP is an integrated CFD solution that includes automated mesh generator, per-processor, solver and post-processor.

If you interested in further information, please feel free to contact me at

Best regards,

ogloth September 25, 2012 05:12


The interesting regions cannot be seen in the pictures posted on the advertising page. How are the boundary layers resolved?

I would assume that an accurate boundary layer resolution is important for the kind of geometry in question.


Maged_CFD September 25, 2012 08:51

Hello Oliver,

I totally agree that resolving (or accurate modeling) of the boundary layer is important for this kind of problem, in addition to a high quality mesh.

The choice between LES/DNS vs. RANS, depends on the time and computational/financial resources available for a particular project.

In addition to the benchmark and validation cases available on our website , we did a benchmark study on flow around the bluff "Ahmed body". SimericsMP achieved very good agreement with measurements compared to the results reported in the following study by Moeller et al. .

I can provide you with more information about this study, if you are interested.


ogloth September 25, 2012 09:41

Hello Maged,

the Ahmed body has clearly defined separation points which might not necessarily be the case for the geometry in question here. Thus you might still get reasonable result even if y+ for the first cell is not ideal.

What is your boundary layer strategy? Do you have a zoom in picture of a resolved boundary layer? Which values of y+ can you reach with your mesher?


JBeilke September 27, 2012 03:42

Doing an optimisation of of such a body is quite difficult. So you might look at an "adjoint" methodology. Have a look at:

All times are GMT -4. The time now is 11:28.