FEA problem finding the good results in deformation

[StJP]

Hi everybody,

I'm a new user of Inventor and my company asked me to analyse their products (doorways in PVC and aluminium) with the stress analysis module of Inventor 2015.

I started by modeling a prototype (a sliding type) we already have mounted in order to compare my results with a concrete case but I cannot obtain the value of deformations I have with the prototype. I put a load of 100N on the top of my model and I consider the gravity.

I have simplified my model and I had the thin body problem so I have made the "find thin bodies" option, my resultats are better but I still am far from the real results.

I have soft springs and my assembly is divided into several independent pieces but the efforts are visually transmitted as they should be.

I simplified the wheels by 2 rotations and contacts sliding/ no separation. I have checked the material several times and even put caracterisitcs lower than them should be.

I have attached several pictures if it can help you.

I would be very glad if you could tell me what could be wrong because I'm on this for several weeks and I'm starting to feel like hopeless.

[StJP]

I alsotried the same assembly without the blocs holding the doorway and the soft springs keep it from simply falling to the ground. I suppose it's the cause of my problem but I cannot find how I could remove them or just lower their stiffness.

[raviburla]

Hi,

Would you be able to post your model (or a simplified one) and the expected results? Perhaps the setup needs to be modified to achieve expected solutions, we will be able to look into those issues if you can post your model.

Thanks,

Ravi Burla (Autodesk)

[StJP]

I've attached my model but the FEA folder was too large. I should have around 200mm in maximum deformation at the extreme top  (where my load is applied).

[raviburla · Accepted solution]

Hi,

Thanks for the dataset. My understanding of your expectations is the following:

* When a force is applied on the door, the door would rotate and the displacement of a point near force application should be 200 mm.

If the above is correct, then I would say that Inventor Stress Analysis is not the proper tool for this type of analysis. Inventor Stress Analysis is a "Linear Static Stress Analysis". This means that the deformations and strains are assumed to be small and model kinematics are supported in a limited sense. For this model, there would be moving parts and a proper kinematic and dynamic analysis is needed.
If the stresses are not important for your analysis, then Dynamic Simulation can be used. This provides a solution with model kinematics as well as forces/moments acting on the joints. However, if stresses are important, I would recommend using Simulation Mechanical with MES, this tool provides both kinematic stress solution.

Please let us know if you have more questions/comments.

Thanks,
Ravi Burla (Autodesk)

[StJP]

Thank you for your answer, in fact I'ld like to study the displacements supposing that we stay in elastic deformations, but I'll have to study stresses as well, so if I have understood what you have written :

- I can study displacements in Dynamic Simulation as long as I stay in the elastic area and I won't be able to evaluate the stresses that way 

- I can study stresses in the stress analysis but my displacements won't be realistic

[raviburla]

Hi,

When Inventor Stress Analysis is directly used for this analysis, then the solution that we obtain would be limited to "static stress analysis". So both the displacements and stresses will not reflect the problem we are solving.

When Dynamic Simulation is used for this analysis, the model would be assumed to consist of rigid bodies. But the kinematics (displacements, velocities, acceleration) of all bodies will be analyzed accurately. In addition, we can also get accurate forces acting on the bodies/joints.

There is a mode in Inventor Professional - One can transfer the loads from Dynamic Simulation to Stress Analysis. In this mode, we would first analyze the model using DS and then choose a time step where we want to understand the flexible behavior. Then the loads(joint forces/moments, intertial loads) corresponding to this configuration (time step) can be exported to Stress Analysis. Once in Stress Analysis environment, the body can be treated as flexible with appropriate material properties and understand the stress behavior. This would be the best approach to get best of both worlds (DS/Stress Analysis). Please note that only one part can be nominated for stress analysis from DS.

I hope this answers your questions. Please let me know if you have more questions/comments.

Thanks,
Ravi Burla (Autodesk)

[StJP]

Thank you