Inconsistent meshing error during static load simulations

[Anonymous]

Hey there,

My name is Steven, a 27 year old master student who is fairly new to Fusion 360. I started using Fusion 360 for my master thesis research. The goal is to simulate 5 different variations (varied by 3D printed layer height and material properties) of the same propeller (APC 11x4,7 SF) under a variety of static loads (50, 100, 250, 500 and 750 grams) to predict practical results. 

I replicated the propeller CAD model as part of my research. The CAD model is sliced with Ultimaker Cura and the G-code of each model is converted to a .stl-file with Voxelizer 2. This file is loaded into Fusion 360 where I did the following in the design environment:

- Reduced the number of facets in the model from the model to 9999. The imported model had millions of facets which Fusion 360 can't compute with. 

- Converted the mesh model to a BREP

- Applied custom material properties to the BREP 

In the simulation environment I did the following (see added screenshots):

- Simplified the middle section of the propeller into a single facet on both topside and backside

- Applied a constraint to the propeller at the simplified topside and backside of the middle section

- Applied a static load to a facet or node at each propeller blade (in the middle at around 9,25 cm from the middle)

- Used pre-check to see if the program has all the information it needs to solve the study.

- Cloned the study and adjusted the amount of force per study to match the desired load

- Solved the studies

The problem: 

I repeated these steps for all five propeller models (see added screenshots). 3/5 Models had no issues and solved fine. At some studies I had to move the force around to avoid getting the meshing error. But with the other 2/5 models (Fail_1 and Fail_2 in the added screenshots) I keep getting the meshing error whatever I try.

I am fairly new to Fusion so I don't know if this is a rookie mistake that has a simple fix. All the options I could think of didn't solve my problem and after one week I am still stuck with the unsolved studies.  

Hopefully, someone out there on the all knowing internet can help me with my simulation/meshing problem. It would be a very valuable addition to my master thesis if I can finish the static load simulations via Fusion 360.

In adcance, thank you so much for your help.   

With kind regards,

Steven 

[henderh · Accepted solution]

Hi @Anonymous with the self-intersecting surface mesh error, it may be bad geometry, and worth using the Validate geometry checking tool (Inspect panel in the ribbon toolbar) in the Design workspace to see if we can detect & fix the self-intersection.

Overall, faceted BReps from STL's are unnecessarily mesh dense, due to the facets (okay for displacements, but introduces stress concentrations at the edges of inside corners, instead of a more realistic swoopy-smooth propeller surface.
Unfortunately, we don't support anisotropic material properties for 3D printing layers, please take caution interpreting the results.
If the loading is symmetric, you can take advantage of symmetry, and split the body in half using the plane of symmetry, and place a Frictionless constraint on the face(s) created by the cut.
This will force the mesher to take a "different route" creating the mesh, and it may succeed.

Hope this helps! Please let us know if you have any additional questions, comments, or suggestions.
Best regards,

[Anonymous]

Hi @henderh 

First of all, a big thank you for taking the time and putting in the effort to help me out. I really appreciate it! Awesometo see that the community helps. 

I know I have to be cautious with the results. For the thesis, I only use the Fusion results to make a gross estimation of what the real test in practice may result in. But thanks for the advice. 

Now, back to the problem: I used the validate tool from the inspect tab as you said. I used the following settings:

- Gap healing = 0.05 mm

- Checked the boxes to allow repair, small edge/face removal and refit bad surfaces.

It detected 64 intersecting or overlapping surfaces, and removed 64 facets and 34 edges. I think that this is indeed an indication of geometry imperfections that originated in the reproduction of the original surfcace.

However, this unfortunately created a new problem. I am now stuck with a body that is not a solid surface anymore (see added picture). It is an bodywith 64 holes in it. When I go to the simulation environment, the 'pre-check' results in an expected error (see added picture). 

I tried to stitch the surface to recreate a solid surface, but I get a modeling error instead (I don't know why). I am concerned that the large holes in the middle section and some additional facets around the propeller surface (marked with red edges, see added photo) are the problem here.

I feel that I am one step closer to the solution, thanks to your help. So I want to thank you again for the tip!

It would be awesome if you could also help me with fixing the arisen gaps to recreate a closed surface again. I hope that once it is a close surface again, this will be enough to run the simulations.

If this cannot be fixed, maybe I should consider your other option of splitting the propeller with the symmetry assumption. But I have no idea how to do that, so my hope for now is with fixing the above problem.

Thanks again for all the help! I hope you can help me with the new problem. 

With kind regards,

Steven