Hello, I am trying to model the flow pattern of a Wankel engine in 3D, In the past I solved the Reynolds equation for a tilting pad journal bearing considering thermal effects using the hybrid FEA method of Baliga, but now in this engine the problem has another proportion so, I started with some books of finite element analysis but for me is quite difficult. I have four mains doubts: 1)I read somewere that FEA does not have(!!!???) false diffusion problems, but in some SAE papers I found that some authors use dissipation funtions to avoid it and they said they use FEA, this mean variational method??. 2)I cant understand how some people include artificial eddy funtions before solving the Navier Stokes equation, could somebody tell me how is this? 3)In this CFD problems, Is the energy equation (i.e. First law of thermodynamic) solved by FEA at the same time of the Navier Stokes Ecuation and continuity equations for a diferntial element? or The space is solved first for continuty, second for N-E, and tree for energy and iterate again until get the smallest error? 4)What mean these turbulent kinetic energy factor and Dissipation rate of turbulating system? Is this the way to include eddy effects on the flowfield?

Please if somebody can send to me a sourcecode in C, C++, basic, or fortran of navier-stokes 3D (well 2D can help!) flow I will be grateful, is very hard to learn this only reading books and shootig to everywhere.

I apologize if I have been too curious,

Yours sincerely....Jose Choy.

P.D. I know that could be easier if I use a comercial CFD software, but I really want to learn how this thing is done! Thanks.

(1). I can only say that to model the wankel engine, you need to solve flow problem with moving boundaryies. (2). It has been done, because I have seen the published results many years ago. (3). What you need is the right school and teacher, to guide you in the cfd research.

Thank you Chin for your help.