What type of Contact best models bolted connections in FEA?

[bmarjaba]

I am conducting a stress & modal analysis on an assembly. I omitted the bolts and corner supports from the study, therefore I need to represent those connections some other way. What is the best way to represent the bolts using Contacts?

I attached screenshots of my assembly along with the different type of bolted connections.

[swalton]

What is your goal?

Do you need to know how the structure behaves near the bolts?

Do you want to know the stress distribution in the thread profile?

Do you need to make sure the mass of the attached components affects the modal analysis?

Answering those questions will help you decide how to build your FEA approximation.

[kacper.suchomski]

Hi

If the analysis concerns the entire structure, it is best to disable the bolts and define connectors in their places (available in Inventor Nastran). This will simplify the construction, reduce the file size and speed up the analysis.

In the case of the strength of the screws themselves (more efficient), do it on a single file, and it is best to use the engineer's guide (also the one built into Inventor).

[bmarjaba]

Thank you for your replies gentlemen @kacper.suchomski @swalton.

As a first step, my objective from the stress analysis is to analyze the entire structure, specifically the parts marked with a red arrow in this attachment, and determine how far we are from yielding with the current material thicknesses, and number of corner supports & bolted connections. In my current analysis, I disabled the bolts, and proceeded with the simulation by representing the bolts by bonded contacts (I think there is a better way, either a different type of contact or choosing the appropriate elements to bond with one another). @kacper.suchomski can I pull off what you recommended in the regular Inventor?

Maybe in another simulation, I'll look into areas around the bolts or stress distribution around the threads.

With regards to the modal analysis, I would like to examine the response of the structure to external vibrations excited by a fan motor resting on the blue rubber isolators/dampeners.

[swalton]

---

@bmarjaba wrote:

Thank you for your replies gentlemen @kacper.suchomski @swalton.

As a first step, my objective from the stress analysis is to analyze the entire structure, specifically the parts marked with a red arrow in this attachment, and determine how far we are from yielding with the current material thicknesses, and number of corner supports & bolted connections. In my current analysis, I disabled the bolts, and proceeded with the simulation by representing the bolts by bonded contacts (I think there is a better way, either a different type of contact or choosing the appropriate elements to bond with one another). @kacper.suchomski can I pull off what you recommended in the regular Inventor?

Maybe in another simulation, I'll look into areas around the bolts or stress distribution around the threads.

With regards to the modal analysis, I would like to examine the response of the structure to external vibrations excited by a fan motor resting on the blue rubber isolators/dampeners.

---

Here are my quick thoughts.

1. I try to balance the FEA setup and solve time with the amount of information I need to build a safe and serviceable structure.  Engineering judgement, code requirements and economics all feed into the amount of simulation required for a design.
2. The FEA in Inventor is limited to some basic assumptions.  If the question that you want to answer does not fit into these assumptions, then you will have to use a different FEA solver to get reasonable results. See: https://www.autodesk.com/support/technical/article/caas/sfdcarticles/sfdcarticles/Stress-analysis-as...
3. Model the bolted connection in more detail if and only if you need to understand the stress distribution around the bolt holes or in the actual fasteners.  Otherwise, the bonded connection is reasonable.
4. Will a traditional hand calculation suit your needs?
5. Take a look at this discussion about ways to model bolted connections if you need more detail: http://wanderlodgegurus.com/database/Theory/FEA%20Bolt%20and%20Bolted%20Joint%20Modeling.pdf
   1. Google shows many different discussions about how to model bolted connections.  It is possible to go into lots and lots of detail.  The design or structural engineer will need to decide how much detail is necessary to address the design requirements.
   2. Some of the methods discussed will work in Inventor FEA, others may require a more advanced solver like Inventor Nastran or Ansys.
6. Inventor FEA assumes linear materials.  Rubber bushings are not made of linear materials.  The Inventor FEA solver may not give useful model results if you need to see the effect of the rubber isolators.  Inventor Nastran or Ansys may be more suited to the physics of your problem.

[bmarjaba]

Noted. I switched to Nastran's environment and will be using it for my FEA to account for the isolator's non-linearity and to use connectors in place of the bolts. I assume there will be no need to use bonded contacts anymore since I'm using connectors right?

[kacper.suchomski]

You can choose from many types of connectors.

https://www.youtube.com/watch?v=dJosjb8yolk

https://www.youtube.com/playlist?list=PLtakexwsjr-BwSo7MiiZckbbgMaYhTioT

[bmarjaba]

I used bolt-type connectors in Nastran, ran the model and got a singularity error at a node located on an unnecessary hole on one of the parts. I removed this hole in Inventor Pro, and when I came back into the Nastran environment, I got a warning saying all my midsurfaces and constraints have been lost: Face has been lost due to Part modification, even though I only suppressed one hole of one part out of 20 parts. I lost all my connectors as well. I then tried to Manage --> Rebuild All, then the software froze and crashed. I tried this multiple times and the same thing happened. Is there a work around for this? Without having to repeat all the connectors, it just doesn't make sense to have to repeat all of them every time a small modification is made.