CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   FLUENT- your opinion (http://www.cfd-online.com/Forums/main/2455-fluent-your-opinion.html)

Chatterjee August 8, 2000 05:35

FLUENT- your opinion
 
Dear Friends:

As my Institute is planning to purchase a PC-based CFD code may I seek your opinion on FLUENT/ PHOENICS and beyond

mehdi ben elhadj August 8, 2000 09:09

Re: FLUENT- your opinion
 
Hi,

PHOENICS is the best.

Parabolic, Hyperbolic Or Elliptic Numerical Integration Code Series.

I use it every day.

John C. Chien August 8, 2000 11:29

Re: FLUENT- your opinion
 
(1). The purpose of the purchase is one of the factor you need to consider. In other words, it it the relationship between your institute and the code. Since you did not specify the purpose, it is hard to make a good suggestion. (2). PHOENICS has been around for a long time, and it has a royal brand of Imperial College school of thinking. It is also easier to find the related publications on the technical journals. I have not used the code directly, but friends around me have used the code many years ago. (then, it was the only code available.) So, this code has closer relationship with the academic institutions. (3). Fluent has gone through two stage of development. In early days, everyone was doing the same thing, the pressure based, structured mesh. So, it was in the same category as the other code. In mid-90's, the UNS for the unstructured mesh beame available and it makes the complex geometry easier to handle. But it also comes with the disadvantage of using the tri/tet mesh near the wall. So, the wall type flow problem could not be handled accurately due to the presence of the wall boundary layer,which requires the boundary layer type coordinates. (4). Thing has gradually changed and improved in the near wall area, and the hybrid mesh was then available. This in principle can solve the wall layer problem, but in reality, hybrid mesh is rather complex, relative to the structured boundary layer hex/quad mesh. (5). So, you can learn different things from different codes. (the disadvantage of using commercial code is that it will give you the illusion the the fluid dynamics problems can be solved by running the code which tends to supress the desire to create something new.) The principle is: if you always follow somebody, you will always lagging behind. Other than that, you are free to look at every or all cfd codes available. you learn more that way.

Evan August 8, 2000 12:19

Re: FLUENT- your opinion
 
We've been using FLUENT on the PC at my company since 1995. We investigated many available programs while selecting a code in 1995, and found that may couldn't properly solve our particular class of problem (buoyancy flows through porous media with small velocities and large length scales). The bottom line is that you have to tailor the CFD code to the intended application.

Kai Kang August 8, 2000 12:56

Re: FLUENT- your opinion
 
It really depends on what you want to use it for. I don't know about Phenics, but I've using Fluent for quite a long time. Fluent runs almost perfect on UNIX workstations, but on PC, I would say about 75-80% as good as on UNIX.

For educational purpose, I would recommend you get the older version of Fluent v4.X as well, which only supports structured grid, and for simple problems, you can generated the mesh and solve all in Fluent (now in Fluent 5, you need Gambit for the model).

Otherwise Fluent5 & Gambit is the choice.


All times are GMT -4. The time now is 00:09.