I'm an automotive engineer studying master in the field of "Improvement of Commercial Vehicles Aerodynamics Utilizing Computer Software" (putting the automation of the processes in mind as much as possible) and I've no experience in CFD, Finite Element Technique or Mesh Generation .

I've chose Ahmed body as the case study I'll work through to learn and join the field of CFD -applied to Automotive - and thought as follows from Tools perspective: CAD: from PTC

Pro/E to get the use its parametric power in optimization

IcemSurf tools to model surfaces and reverse engineer others CAE: from Ansys

Ansys FLOTRAN as Pre&Post processor and solver

(because its available in a multiphysics environment which I intend to use in the future and also because there is a well established connection between Ansys and ProE) but I got the feeling from the discussions on cfd-online that I should use FLUENT ,so I changed the plan to using: CAE:

CFD solver: FLUENT

Pre&Post processor: ICEM CFD because I heard that it has a good connection to ProE /FLUENT and almost all CAD and CAE tools

, so I thought you may have the time to tell me: 1) your opinion about the tools I intend to use (and you may recommend others ,as you see suitable). 2) the prerequisites (in terms of books or papers I should read) I should have to attack such kind of problems with such kind of tools ProE&IcemSurf/ICEM CFD/FLUENT or any other tools you recommend.

thanks for your time Marwan

Your approach seems to be very "academic" and far away from the real problems in car aerodynamics using cfd.

To get a first impression about the standards in this topic you might look at the EZ-Aero and Underhood articles at www.adapco-online.com

Hi,

I think you software selection sounds good.

I can comment on experience with the PROE ICEM CFD link. This connection will make the mesh generation very clean and tidy as you will not have to use IGES data! From this you can make either TET / PRISM meshes for complex models or HEX meshes for simpler models.

For the solver you can either look at STAR or Fluent (both of these are supported by ICEM meshing).

Regards

David

Well, I agree with your case selection. It is a good case to study what is important and what is less important. I do not agree that you have to deal with a real complicated car geometry would help; it can actually hide a better undertanding of the physics.

As a start, try reading a few books and papers on fluid dynamics (Navier-Stokes Equations)and CFD. The reference section on this forum can act as a good start to select some books and works. The most commonly used and most popular approach is Finite Volume Method. I would suggest that you read about that too, and how it works for low speed flows.

You then also have to read about how turbulence is "modeled". This will have a great implication what kind of mesh you would find appropriate for your case.

I can go on and on. But let me stop here.

Well,I am studying the automotive aerodynamic by CFD. I am just using the PHOENICS. Don't know it is correct or not.