Coincident Faces causing issues with simulation solving error.

[Anonymous]

Hi,

I have been working with Fusion 360 over the last couple of months and I have come across an error I can't get my head around. I have assembled a bracket which has a simple socket and pipe arrangement. However when I go to simulation to run the analysis there is a warning regarding interference. The only interference is coincident faces and this results in the simulation stopping due to an error. 

Sorry if I don't understand something fundamental however how can a coincident face be causing a problem during simulations. And the other obvious question is how do I overcome this problem? 

Thanks in advance for any help that can be provided. 

Timm

[John_Holtz]

Hi Timm,

It would be helpful if you can attach your model so that we can see exactly what is going on. For example, what is the message given during the analysis? ("File > Export" and save to your hard drive as a .f3d, then attached the .f3d file.)

One thing to keep in mind with round geometry (a pipe) sitting in or passing through a hole (a socket) is that the simulation is using an approximation of the solid model. The mesh between the parts can be interfering even though the CAD geometry has coincident surfaces, or in an extreme case, the CAD geometry can have a clearance. Here is an example of a pin passing through a hole (see attached "overlapping mesh.png").

If the mesh on the pipe is much finer than the socket, you may be able to avoid the problem by using a similar mesh size on both parts. Either uncheck "Manage > Settings > Mesh > Scale mesh size per part" so that the mesh is the same size on all parts, or use "Manage > Local Mesh Control" and specify the same mesh size on the two parts.

Hope this helps.

[Anonymous]

Hi John,

Please find attached model as requested. Interestingly I couldn't export as .f3d so I have exported as .IGES - it referenced a warning about not supporting .f3d export with referenced components. The error that comes up during solving is: 

Error: FATAL ERROR E5004: STIFFNESS MATRIX SINGULAR OR NON-POSITIVE DEFINITE

In addition, I tried your recommendations below and the 'Scale mesh size per part' was already unchecked.

Thanks in advance for your help - looking forward to the response.

Timm 

[John_Holtz · Accepted solution]

Hi Timm,

The error 5004 is usually not related to any interference in the model. It is trying to indicate that the analysis is not statically stable, so some part or parts want to float off into space. It is equivalent to dividing by 0. (For example, if the model were not constrained in the horizontal direction, the analysis is calculating the horizontal displacement = horizontal force/0.)

You may be able to solve the problem with one of these ideas:

1. Make sure that you have constraints on the model that prevent it from moving in all three directions (UX, UY, UZ).
2. You can try activating (checking) the option "Manage > Settings > General > Remove rigid body modes". You may see a part that is moving strangely if you are able to get results, so this would help to focus on what part is not restrained properly.
3. If you are using a contact type other than bonded, change those contacts to bonded to see if that eliminates the error 5004.

I took a look at the .sat file that you provided. However, that format does not have any of the simulation setup, so I was not able to see how you set it up (loads, materials, constraints, and so on.). I did apply some constraints and a load just to see what it would do, and was able to get results. If you are still having problems, you can export your model to a .f3z format using the instructions on this post: Saving Assembly as .F3D

[Anonymous]

Thanks John,

That worked perfectly - the error was associated with not constraining the model correctly. Everything is working well again.

Thanks for your quick response.