Hi Guys,
I'm performing a Transient Mass-Transfer simulation, and I noticed something strange. Take a look at the following images:
The image on the left shows the results from the simulation. The image on the right is the same part selected in the tree so that all the nodes are highlighted. Say WHAT?! Look at all those nodes that are outside of the physical part!
It's not just in the results environment. Here's the same part in the FEA Editor:
I can even select those weird nodes:
Any idea what's going on here? Is that a graphics artifact, or are my results being affected by these weird nodes?
NOTE: This is one part in an assembly. The rest of the parts are hidden... Maybe it's showing nodes from those parts??
Hi,
Can you select the elements(Selection tab>Select panel>elements)? If yes, then perhaps you can right-click>Inquire element information to find out which part it belongs to.
Regards
Ilyas
@Anonymous wrote:Hi,
Can you select the elements(Selection tab>Select panel>elements)? If yes, then perhaps you can right-click>Inquire element information to find out which part it belongs to.
Regards
Ilyas
Yes, I can select elements, and it says that the element is part of the part shown, even though it appears to be outside of the part itself.
If it says that there is something there, then I would believe that something is there, and it might be affecting your results. What does Result Options>View>Element Visibility>Show All show you? Are the mysterious nodes overlapped by other parts?
Yeah, from what I can tell, the mysterious nodes do overlap other parts... But there are no results shading on them, as shown in the images above. So it's kinda wierd... If they were really there, why aren't they shaded? Hmm.
AstroJohnPE, do you have any ideas?? Is this just a graphical UI issue, or is it something more fundamental?
Because the nodes are shown in BOTH the Results and FEA Editor, I think the software believes there is a mesh there. By default, both environments do not "show" the lines/elements that are "inside" the part. My guess is that the internal mesh has somehow gotten outside of the physical part. Did you do anything unusual, such as creating or changing some of the mesh by hand? (In the FEA Editor, the internal mesh can be shown by "View > Visibility > Object Visibility > Internal Mesh". In the Results environment, "Results Options > View > Show Internal Mesh".)
If the problem were only in the Results environment, I have seen two cases (out of thousands) where the rectangle selection selects nodes in a different part of the model than where you drag the rectangle. But this is not the situation here.
Yes, when I show the internal mesh, I can see all the extra stuff outside of the part both in the FEA Editor and in the Results environment. Here it is in the FEA Editor:
No, I didn't do anything unusual... I brought the part in from Solidworks and meshed it. No manual operations. I adjusted the mesh size on a few of the parts to ease the meshing process, but that's it.
I do not have any ideas how it happened, so I do not know if this will "fix" it or not. My suggestion is to copy the design scenario to a new one and mesh the model again. You do not need to run the analysis to see if it fixes it; either display the internal lines like you did, or do an "Analysis > Analysis > Check Model" to see what happens in the Results environment.
Hi John,
I copied it into a new scenario and I'm remeshing. There's something interesting with the meshing engine, because the first time I tried to remesh, one of the parts failed during tet generation and reverted to using the "old method." Three days later, it was still meshing. I killed it. Then, when meshing again, another part failed during the solid meshing, "failed to solid mesh and/or was not watertight." This part was just successfully meshed in the previous attempt.
Is the meshing engine non-determinant? Is there a random seed generated within that would result in different different results each time I mesh?
Thanks,
Todd
So the solution would not mesh in Design Scenario 2 using the same parameters as in Scenario 1. I had to reduce the mesh size of one of the parts to avoid that tet mesh failure using old method error. Once I did this, the solution meshed completely.
The internal mesh is visible outside of the part again, similar to, if not exactly the same as, Scenario 1.
Can't find what you're looking for? Ask the community or share your knowledge.