CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Turbomachinery (

MSc February 15, 2011 15:47

Hi, i'm fairly new to the CFD world, i'm doing a study on a very large axial cooling fan in South Africa. I'm looking for direction in using Ansys, specifically Fluent with regards to turbomachinery?

How would I go about modelling the fan?


red_lemon February 16, 2011 18:45

The level of your question can only mean one answer.
Employ someone else to do it who has the skill and experience.

MSc February 18, 2011 08:06

I obviously didn't state my problem correctly. I'm conducting a STUDY on this. Therefore there has no work been done previously that we know of with regards to the modelling of an axial fan of this size.

I would like to know what kind of tools (if any) Fluent has to help with the modelling of this fan or turbomachinery in general.

red_lemon February 18, 2011 08:41

ok now I understand a bit better.
ANSYS has geometry and meshing tools like ICEM, blade modeler, Turbo grid etc that can be used for this type of thing. ICEM Hexa could be a good tool for a more explicit modelling approach. Then best to use CFX for the solve part and CFDPost at the end. You can start with an implict approach with pressure rise and swirl then move to fixed blade explicit model and then to MRF then transient sliding mesh. You can make use of rotationally periodic zones (cyclic symmetry) too to reduce solve time.

Do you have a CAD model of the fan to start with? or even a fan curve? If no geometry could it be built in CAD or ANSYS geometry engine? or even scanned with a laser to generate surface profiles for the blades?

MSc February 21, 2011 09:02

Thank you for the reply. I only have FLUENT to my disposal at this stage. It seems though that Turbo grid for example is only available if using CFX? Is this true? Is it worthwhile trying to get hold of CFX?

Atze February 21, 2011 10:46


I use Gambit to make meshes for Fluent. If your geometry is simple you can simply use it, else you can import it from a .iges .igs. Once you have your mesh in fluent you can analyse it with MRF (steady) or MM (unsteady) models.

MSc February 22, 2011 06:40

That's good news. I have the .igs files of the blade and I was wondering how one imports these files into the ANSYS workbench. I use version 12. Is it as simple as you made it sound?


MSc February 22, 2011 06:45

red_Lemon, I'm considering using bladeModeler. The question I have is if one can use bladeModeler to not only create a mesh for the blade itself but also for as in my case the bellmouth and windtunnel casing around the fan?

Atze February 22, 2011 10:10

Yes, it's pretty simple. In gambit you have just to import geometry from iges file. Import only edges, because sometime Gambit imports corrupted surfaces. Better if you import (from a .txt file) the verteces of your airfoil, and combine them with a spline.
The point is that Gambit (sometime) can't read .iges perfectly (virtual edges, degenerate surfaces or volumes) so you need to fix it. But, i repeat, if your geometry is not so complicated, you should not have troubles

All times are GMT -4. The time now is 20:15.