Hi everybody,

Well my supervisor told me that i have to make some modifications to CFX if I want to continue working with it for my PhD.

My question is this is possible and how should I start for this? Do you guys have experience with this?

I am not sure what exactely I have to do but I will be back soon with details.

caty

Where will you get the source code from ? Its a dauntung task, perhaps you should find a program that is more suited for your research than trying to modify one, but then it depends on how extensive your planned modifications are.

Why not write some user fortran to do something novel or clever with CFX?

You don't need the source code. You can make lots of changes/additions by cel or user fortran. It all depends on what you want to change (you have to be more specific). You can yourself do it or have CFX do it for you (at a price). I dodn't see a lot of exchanging of CEL or userfortran between users yet (not for CFX 5 anyway). But if you state your actual problem here someone might help you. Other thing might be that your professor might ask you to do this because you want to be a PhD and you cannot simply get pretty pictures from a commercial code and get your title (some do this by the way)

I think you are all partly right, it depends on her area of research, if you want to change something fundermental as coming up with a new or modified turbulent model for example you may need acess to the code or some entry point for including your own work, if it is just a slight modification then a fortran rutine(s) will surfice. I have some friends doing their Ph.D, and its tough without the support a big company like Ansys-CFX will provide, and even then they only concentrate on a small area of improvement for a particular problem.

Hi,

In my experience user fortran is rarely needed. Many people think they need to write user fortran when what they are trying to do can be done as CEL expressions. Maybe I just do simple stuff but I don't even have a fortran compiler.

Of course, a PhD is neither using a commercial code or even writing your own code. A PhD is a "contribution to knowledge" and CFD is just a tool to allow you to understand fluid flow better, whether you bought it or wrote it.

It all comes back to what you are trying to do. Caty, what model are you trying to add to CFX?

Regards, Glenn Horrocks