Rigid Connector proper application

[oharag11]

I'm using SW Sim Pro and trying to run a modal analysis of a hanging frame structure. There are 3 faces that I would like to connect using a Rigid Connector because a semi rigid object will be attached to these faces using fasteners. When I enter into the Connectors - Rigid Connection setup there is a pull down for different connections. If I choose Rigid there are two selection boxes (one blue the other pink). If I hover over the blue it says - Faces for rigid connection, and the pink says - target. I tried looking for an example that mimics my setup, but the info is very generic without examples. So, I believe I would select the three faces that will "move together" due to the hanging mass - but what do I select for Target?

 

[FEA way]

From the documentation:

A Rigid connector rigidly connects faces from one solid body to faces from another solid body.

So you should select faces from one solid body as "Faces for Rigid connection" and faces from another solid body as "Target".

 

[oharag11]

FEA way,
This seems appropriate for let's say a circular rod that spans two surfaces. I have three surfaces - all tied together via an object fastened to frame. Should I be using
Links??? I guess I could use a Fixed - Fixture for all three surfaces - but I do not believe this to be accurate since all three surfaces should move together and be semi-rigid during translation.

Maybe I'm assuming/using Rigid Connection incorrectly? Again - terrible examples within SW help and the internet.

 

[FEA way]

Can you share a screenshot of your model or a simple sketch ? Maybe multiple rigid connectors could work.

 

[oharag11]

I always find these - attach a sketch requests kind of funny. I tried my best to mimic my setup. I work in an industry that protects IP so no CAD screenshots on my end.

If you open the jpeg you will see black box (item being suspended within a frame - somewhat semi-rigid) - RED rectangles - support structure to attach box to during suspension - front - top and rear. and Green lines representing a frame to support box.

The black box adds some sort of structure/rigid frame to the enter supporting frame. I want to mimic this support for Modal Analysis to see impact on modal frequencies. We will be shaking this frame/black box and I want to mimic real world as well as I can within SW Sim.

Here's how I would expect Rigid Connection to work in this scenario - It would allow me to select as many surfaces I need (2, 3 or more) and then SW would assume these surfaces move together. If I fix each of the surfaces (maybe constraining them individually) I believe this doesn't represent the rigid hanging box as it is mounted within the frame. I guess I could use Links between the mounting holes for each of the three mounting surfaces - but I would create many linkages (4/5 mounting holes per mounting surface). I could create a remote load (weight of system) and tie each connection hole together using a rigid connection.

So, I ran the sim again using two rigid connections - one from the rear (faces for rigid connection) and the front (target) and then another from the top (faces for rigid connection) and the rear (target). I do not know if this conflicts with each other or causes any issues. Modal analysis did not fail. Looking at the shapes it does look like they are moving together - ie. Mode 1 shifts right - and all three surfaces shift same amount. Looking at other mode shapes it's hard to determine if the surfaces are acting accordingly.

I hope this helps.

PS Do you see an attachment - I don't on my end and I uploaded and made attachement.

 

[oharag11]

Attachment

 

[FEA way]

So the black box is the part that you don't model and only want to account for its presence, right ? I would try with overlapping rigid connectors like you did. The problem with CAD-embedded FEA modules is that they hide a lot of important details from the user. Here, it may not be easy to determine whether there is an overconstraint or not. Many FEA solvers can resolve some overconstraints automatically but not all of them. The selection order (source and target) may also matter if it's implemented as a traditional master-slave constraint internally. But if there are no errors and warnings about overconstraints then it should be fine.

 

[oharag11]

FEA Way

Yes, black box is not in SIM - way too complex to mesh. The goal is improving the mounting structure so that it won't contribute to the black box during Vib testing.

Weird thing FEA Way - if I add loads to the mounting holes to mimic weight of black box - I cannot use an iterative solver (i.e. FFEPlus). I have to downgrade to the direct Sparse Solver - and when I do this I cannot use "Upper bound frequency" as an option. Do you have a clue why this is the case? It would seem users would like to perform Modal Analysis of a system (i.e. adding fixtures and connections to accurately mimic the system - and even add a weight load of the system) to get accurate results.

 

[oharag11]

BTW in one scenario I added all three attachment surfaces in the blue box - and the floor of the fixture as target. The floor is fixed using a constraint. I do not remember results, but I may rerun this scenario. Again, the floor is fixed during sim.

 

[FEA way]

What kind of load ? Forces can't be normally used in modal analyses (only for preceding preload). Only additional (e.g. remote) masses can be applied.

 

[oharag11]

I added a downward force equally among all attachment points representing the total weight of the system. I do see there is a distributed mass load so maybe that's more appropriate. I'm rerunning as I type. I'm also going to modal a black box - attach it to frame - give it a given mass - but make it stiff - and then rerun analysis. I will also attempt to do a remote load to compare results.

I will also run a structural Sim with load, and then share this with modal analysis to see the impact. I will give this a try. I do see you can add thermal/CFD and structural results to a frequency analysis.

Thanks