Which one is better??? and why? whats the difference?

4.3 is only for structured meshes and has a lot of models in it (allows for user coding). 5.3 is only for unstructured meshes and has a VERY limited number of physical models in it and does not allow for user coding (not even in version 5.4)

Could you please elaborate more on 'very limited number of physical models'? Thanks!

not available in CFX 5.3: cavitation, mass transfer between phases, free surface flows, extended turbulence modelling (only basic models are available), general user coding (fortran routines), real gas modelling, moving grids etc.

Look at the CFX web site. I myself work witrh STAR-CD 3.1 and CFX 5.3

Major advantage of CFX 5 is the robust and fast coupled solver (they got it from tascflow but did not transfer the modelling yet available in tascflow)

Also,

- Parallelisation in CFX-4.3 stinks. 5.3 and 5.4 are far superior, scalable to as many CPU's as you want as long as you increase the grid size.

- 5.3 and 5.4 can do structured meshes BTW. You can import them meshes from a number of third party grid generators, as well as CFX-4 and TASCflow.

Hi, Like the other people have said the main differences between CFX4 and CFX5 are the structure and unstructured approaches to meshing. If you are looking to model a complex geometry, requiring little in the way of additional user added physics then CFX5 is the one for you. However if the model is relatively simple but you want to implament say your own turbulence model, then you would want to do this in CFX4. CFX are hoping to bring out next year CFX5 with user fortran, which will then allow much greater user interation with the code. Version 5 is the way forward, but its still got a little catching up to do when compared with 4

5.3 is a 2 year old version of CFX-5. Quite a few more features have been added since then. One might also point out that CFX-5 can run on a structured mesh just as easily as an unstructured mesh if you use the mesh import features (CFX-4 meshes obviously being one of the choices). So, if your modelling requirements are within the realm of that which CFX-5 handles, CFX-5 is the better choice.

Did you know that in the latest version of CFX 5.4 the ratio Cp/Cv is set to 1.4 whatever you do? They lost a lot when going from 4.3->5

Your statement is not correct. You can change the ratio of specific heats using the expert parameter:

specific heat ratio = gamma

Gamma is only used when calculating the total pressure for an ideal gas.

Dan.

If you want to model real gas effects a constant gamma is of no use. The calculation for total pressure is wrong anyway (is calculated assuming a ideal/perfect gas)

Yep. You are right. CFX-5 can properly deal with constant density or density defined by the ideal gas model. I don't belive that anywhere in the documentation it says that CFX-5 can model real gas effects.

Dan.

Dear Sir,

I have found CFX 5.4 is very convelient to create Furnace Geometry. But the problem is CFX5.4 has no facility to simulate Coal Combustion.So, if I like to import CFX5.4 geometry file by CFX 4.3 for Combustion, is it possible and how?

Regards,

Monir

It's impossible now. Because the cfx4 is structure mesh and cfx5 is unstructure mesh.

Thanks for your clarification. Regards,

Monir