Error in CFX Pre - Meshing Archive Not Existing

  • Last Post 11 February 2019
fad6995 posted this 06 February 2019

Hello, kind regards to the Ansys Community.

For several days, I have been trying to simulate the air flow through this hood expecting to accuaretely determine the pressure drop. At first I try simulating an imported geometry from Solidworks, then I substracted this geometry into a cubic domain and finally I try modeling the hood in Design Modeler; all with the same result (or error). Once the geometry complement is checked, I logically move on to the meshing, in which the automatic meshing throws an acceptable skewness. Everytime I try to open the "Setup", this error appears:

Error message

The hood geometry is actually not that complicated, reason of why I fail to understand the origin of this error. For the geometry modeled in Design Modeler I decided to use a Simmetry condition, so it visualizes as it follows:


Hood geometry


I pretend to apply a border mass flow condition to one end of the duct, then an opening condition to the hoods, and last but not least, the simmetry condition at the longitudinal cut. Nevertheless, in the "Setup" after the error message pops-up I can only visualize this: 



As final notes these are the meshing characteristics and I am using CFX instead of Fluent for beginner's choice as it seems friendlier to work and the computational requierements that (as I have investigated) are supposed to be lower.

Average skewness: 0,515.

Thank you for your time and support, sincerely




Attached Files

Order By: Standard | Newest | Votes
rwoolhou posted this 07 February 2019

Not sure about the memory need: I've tended to find they're marginally in Fluent's favour but once you start adding models there's not much in it. Whilst both codes currently have similar functionality I'd suggest Fluent is the better option simply because of the user level in most industries. I've used Fluent for a long time, and can use CFX  at graduate level. 

Please review the various mesh tutorials & you tube videos: the mesh you've shown there is far to coarse, even to the extent where the shape is no longer well represented. I agree the skew is good, but there's a lot more to it: CFD isn't a quick subject to master. Turn on curvature & proximity in the panel (bottom image) and compare the meshes. 

The mesh may not be the cause of your problems, but equally the above won't run well. 

fad6995 posted this 10 February 2019

Rwoolhou, thank you for reviewing the case and reply to it.

Indeed, it is not quick and I happened to have rushed a little bit into this simulation. I will take a deeper look in the mesh.

On the other hand, the problem was solved by changing the folder where the archive was beed saved, maybe it had something to do with the original folder was auto-uploadable to the cloud. 

Kind regards.


rwoolhou posted this 11 February 2019

Pass - however that may have corrupted something if it started uploading during the save process. 

Take a step away from the computer and sketch out the domain on paper, then write down all of the boundary conditions, materials etc. This helps decide what mesh resolution is needed and also highlights which models to activate and, more critically what you don't know. Then move to the computer and rebuild your mesh & carry on into solver etc.