Simulation Mechanical Forums (Read-Only)
Welcome to Autodesk’s Simulation Mechanical Forums. Share your knowledge, ask questions, and explore popular Simulation Mechanical topics.
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Symmetry results problem in Autodesk Simulation Multiphysics 2013

5 REPLIES 5
Reply
Message 1 of 6
JDMather
1822 Views, 5 Replies

Symmetry results problem in Autodesk Simulation Multiphysics 2013

We were trying to reproduce results from Inventor FEA.

First we launched directly from Inventor bringing in the settings from Inventor.

The loads are symmetrical, but analysis results of stress and displacement are not?

If I increase the mesh density the results look closer to what I expected but takes much longer than in Inventor.

We used all tetrahedra for the mesh.

 

Can be reproduced with the attache STEP file.

Fixed constraint on cylindrical shaft.

200,000 lb load on each side parallel to the axis of the cylidrical shaft (-y).  (View>Shaded with Edges to see the faces where load was applied)

 

Any comments?

 (I realize we should remove the small fillets and holes.)

 

Edit - after examining the default mesh I see that it is not nearly as symmetrical as I might have expected.

Displacement.png

 

 


-----------------------------------------------------------------------------------------
Autodesk Inventor 2019 Certified Professional
Autodesk AutoCAD 2013 Certified Professional
Certified SolidWorks Professional


The CADWhisperer YouTube Channel


EESignature

5 REPLIES 5
Message 2 of 6
hxq0707
in reply to: JDMather

I think you are not supposed to use "All tetrahedra" in this analysis, just use the default "Bricks and tetrehedra" with a little finer mesh, for example 50%, you will get nice and symmetric results (see my attachment).

All-tetrahedra mesh in Autodesk Simulation is mainly for fluid flow analysis. For static stress analysis, All-tetrehedra will require significantly more elements to get the same accuracy as the “Bricks and tetrahedral” option.

Message 3 of 6
JDMather
in reply to: hxq0707

The reason I used all tetrahedra is that we run the identical (as much as possible) analysis in Inventor, SolidWorks and Creo (formerly Pro/E) and I think those all use a tetrahedra mesh.

 

Our next step is to properly constrain the large hole so that it does not distort in the manner it does in this initial test.

There is a pin that fits in the hole.  I would like to run the analysis without the pin, but will add another part if I can't figure out a way to properly constrain.


-----------------------------------------------------------------------------------------
Autodesk Inventor 2019 Certified Professional
Autodesk AutoCAD 2013 Certified Professional
Certified SolidWorks Professional


The CADWhisperer YouTube Channel


EESignature

Message 4 of 6
JDMather
in reply to: JDMather

We ran the problem today in class with the small fillets, cuts and holes removed from the ends of the hooks.

With slightly finer mesh (all tet) we got expected and equivalent results as seen in the other programs.


-----------------------------------------------------------------------------------------
Autodesk Inventor 2019 Certified Professional
Autodesk AutoCAD 2013 Certified Professional
Certified SolidWorks Professional


The CADWhisperer YouTube Channel


EESignature

Message 5 of 6
hxq0707
in reply to: JDMather

In Autodesk Simulation, those small fillets, cuts and holes cause watertight problem in All-tetrehedra meshing if you don’t use very fine mesh size. That’s why you can get good results now when you have removed them.

When you mention other programs, did you try Ansys?

Message 6 of 6
JDMather
in reply to: hxq0707

I also got good results with the small geometry in Autodesk Simulation if I set the mesh very fine, but it took about 7 minutes to solve.

 

It was obvious I should remove the fine details that had no real effect on the parameter that I was analyzing.

 

We have Inventor, Autodesk Simulation, SolidWorks and Creo (Pro/E Mechanica) here, no Ansys.

 

We go through the Algor University Curriculum running all the labs through all 4 programs that we have.

We get equivalent results in all (as much as possible that we are able to set up the analysis equivalently - sometimes it is not possible to get exactly equivalent constraints).  One thing we noticed - these problems (the Algor curriculum) seem to be set up to validate hand calculations (or rather to validate that the software returns results similar to the hand calcs) rather than real world problems.  It would be nice to have a complete set of real world problems.


-----------------------------------------------------------------------------------------
Autodesk Inventor 2019 Certified Professional
Autodesk AutoCAD 2013 Certified Professional
Certified SolidWorks Professional


The CADWhisperer YouTube Channel


EESignature

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk Design & Make Report