hi everyone,

I´m pretty new on fluent and i´d like to know wich solver and boundary conditions will work best for automotive external aerodynamics,i saw at many articles that i have many usefull options for that task.thanks in advance

Hard to say, but for reversed flow, seperation etc, on automotive surfaces, Realizable is recommended. There is plenty of literature available, you just need to go find it. k-epsilon predicts delayed separation, and flow appears to be attached, when in reality it would have separated. In previous work i had done, I found that RNG gave best predictions, when compared to experimental, for my specific problem (prediction of seperation on curved surfaces).

Since you have reviewed the articles you have the answers you are looking for. Afterall, the boundary conditions applied in this class of problem are Dirichlet bc's. Go and re-read the articles or at least read the FLUENT Users' Guide, which has a useful introduction to the many bc's you can posisbly use.

first of all,thanks for answer.

regarding my post,what i want to know is about personal experiences or recommendations for this kind of problems,since fluent user´s guide doesn´t says anything about this,i thought i might save some time asking for suggestions instead of experimenting.thanks a lot. if you have any other suggestion,i would lilke to hear it.

hallo leo,

for external aero both RNG and realizable are a good alternative, but if you have enough resources in your machine i suggest you RSM (at least I order). But pay attention: it needs a very good mesh and it is much more expensive in time consuming. So if you are new in preparing model for CFD i think RNG could be your starting point.

thanks for your suggestions Davide,but as you said i´ll take the RNG option since i don´t have neither the resources nor the expertise on CFD. any suggestions on turbulent initial B.C´s? wich turbulent lenght/turbulent K.E. shall i take?(or if you choose any other turbulent B.C.,please let me know)thanks again!

i usually use "intensity/viscosity ratio" option: new wind tunnels have a Tu=0.5-0.2%. i suggest you, if you can, to make at lest two (/three7) test. moreover make sure that your test chamber has a blockage less than 5%, so you are sure that walls don't influence your domain.

i hope that this could help you.

actually it help a lot,thanks!!

Seems to be that you have a lot of experience on this kind of simulations,if you would like to give another usefull advise (like those of the above)it will be welcome.you can mail me if you want Sorry about the world cup,it just wasn´t fair.

Arrivederci!

good luck then and feel free to contact me whenever you want!

Davide.