Learning Forum

[acwh]

Hello ANSYS community

I am trying to simulate the fluid flow over an aircraft but I am constantly getting a float point error despite refining the mesh. I have also tried to group the whole aircraft assembly as a single part, both hybrid and standard initialization but the float point exception still persist. 

Attached are the mesh and set up details for reference and I hope someone could help me out!

Thank you!

Figure 1: Mesh Details 1

Figure 2: Mesh Details 2

Figure 3. Mesh Details 3

Figure 4: Mesh Details 4

Figure 5: Set up details 1

Figure 6: Set up details 2

Figure 7: Set up details 3

Figure 8. Set up details 4

Figure 9. Set up details 5

Figure 10. Set up details 6

Figure 11. Set Up details 7

 

 

[Rob]

Look at mesh quality: you've got several flat (very squashed) cells.  Ortho quality should be over about 0.1 and skew below about 0.9-0.95 depending on cell type & flow gradients. 

[acwh]

Hi Rwoolhou

Thank you very much for the reply.

May I check if there are any ways that I can overcome the skewness issue?

[Rob]

There are. First off you need to find the cells (click on the graph bars of the poor cells) and work out why they're bad. From there you eliminate the cause.  Sounds simple but it can take some doing. 

Once you've found the poor cells post images and we may be able to offer advice or (much better) comment on your ideas. 

[acwh]

Hi Rwoolhou!

Thank you very much for the help!

Sure thing!  I have attached the screenshot of the orthogonal quality and skewness quality for your review.

Figure 1. Mesh of Aircraft

Figure 2. Orthogonal Quality (ps: I have trouble showing the orthogonal quality on the aircraft body so I combined two screenshots together and reduced the opacity of the pictures to get the figure below)

 

Figure 3. Orthogonal Details

Figure 4. Skewness Details

 

[Rob]

Now open up the geometry tool and have a very careful look in those regions. My guess is you've got some small gaps/slivers in those regions: you'll need to get rid of them. 

[acwh]

Hi Rwoolhou

Thank you very much for the reply!

I have managed to improve the quality to 6.318E-2 by creating an square box in between the gaps and unite it with the aircraft to close the gaps. I will continue to work on improving the gaps. However I can't seem to update the mesh now as the software will say that "The Process suffered an unhandled exception or ran out of usable memory". I can't seem to highlight the problematic geometry also to isolate the problem.

May I know how can i troubleshoot this problem?

Attached is the screenshot of the meshing stage for your review.

Figure 1. Orthogonal Quality

Figure 2. The Mesh when I tried to update the geometry

Figure 3. The Error Message

 

[Keyur Kanade]

Can you please exit workbench and reopen it. 

The min ortho quality is 0.06. I think this should be good enough to proceed in Fluent. 

Regards,

Keyur

 

If this helps, please mark this post as 'Is Solution' to help others.

Guidelines on the Student Community

How to access ANSYS help links

 

[acwh]

Hi Keyur

 

Thank you very much for the recommendation.

 

I have tried to restart the software but the problem still persist. 

 

May I know if there are any ways that I can overcome this problem?

 

[Rob]

I suspect you've exceeded a face cell limit (there's a cap on the number of facets) if you're now meshing the gaps with sufficient resolution to make the model work. 

[acwh]

Hi Rwoolhou!

Thank you very much for the reply! I appreciate it a lot!

It is weird but I could update a mesh with more elements before I duplicated the file to improve on the geometry gaps.

Attached are the screenshots of the difference in number of elements for your review.

Figure 1. Original Copy

Figure 2. Duplicated and Improved Geometry

[Rob]

Not sure: check size functions and see if you can increase the minimum sizes.  

[acwh]

Hello Rwoolhou!

 

Thank you very much for the help!

 

My apologies for the late reply as I was trying to solve the mesh issue. I have managed to update the mesh after I apply patch independent method to the geometry.

 

 

[Rob]

If that works you've got a sliver face or short edge somewhere. Patch independent just ignores those. 

[Keyur Kanade]

Patch independent is top down approach. It ignores small features automatically. So your geometry must be having small features and that is why patch dependent mesh fails. 

Please note that as you are trying to solve external flow, recommend to cross check if any important geometry features are ignored by patch independent meshing. Then proceed to solver. 

Regards,

Keyur

 

If this helps, please mark this post as 'Is Solution' to help others.

Guidelines on the Student Community

How to access ANSYS help links