How to find deformation of loaded wheel on rigid surface?

Tim2746083295
Tim2746083295 Altair Community Member

Description:
Hello everyone, I am working on a project where I must design the internal structure of a 3D-printed tire. I want to simulate the deformation of the tire under custom loads.

An extremely simplified version of the wheel is shown in the figure. The wheel is constrained so only vertical motion is allowed and a vertical force is applied on the center ring of the wheel. An impact motion contact is defined between the wheel and ground.

image.png

Due to the fact that the deformation will be extreme I was also wondering whether Inspire is capable of solving this, or whether a different program will be needed

Product: Inspire 2024

Tagged:

Best Answer

  • GTT Adam
    GTT Adam
    Altair Employee
    Answer ✓

    Hi @Tim2746083295,

    For the Motion analysis, you will need to convert the cylinder to a Flexbody in order to allow the body to be deformable. The compression of the body can then be viewed using the Flexible Body tool:

    image.png

    If you are experiencing penetration of contacts, then they likely are not stiff enough. You will need to review the contact parameters and increase the stiffness values.

    It looks like the contact size in the Structural Analysis does not dynamically update, hence the deformed shape.

    I quickly mocked up an example using the Motion Analysis. Please see the results below:

    inspire_wheel_compression.gif

    This should be suitable for a preliminary analysis, such as to find out if there will be spoke contacts or general compression amounts. The simulation can be improved using smaller mesh sizes, stiffer contacts, etc.

    If you are trying to use this model for things like durability or lifecycle analysis, I would still avoid using this Flexbody method as flexible bodies are not suitable for large-scale deformation, hyper-elastic materials like rubber, and local plastic deformations, as the superposition of modes only applies to linear systems.

    Hope this helps!

    Adam Reid

Answers

  • GTT Adam
    GTT Adam
    Altair Employee

    Hi @Tim2746083295,

    If you are looking to apply a load to the bottom of this wheel and see the deflection response, a Structural Analysis in Inspire can be suitable to your needs. The 'road' can be replaced with either a distributed force or point load and the results from the various load cases can be analyzed.

    image.png

    I would try to avoid the Motion Analysis and flexible bodies for this analysis, however. It should be noted that flexible bodies are not suitable for large-scale deformation, hyper-elastic materials like rubber, and local plastic deformations, as the superposition of modes only applies to linear systems.

    Hope this helps!

    Adam Reid

  • Tim2746083295
    Tim2746083295 Altair Community Member

    Hello Adam,

    Thanks for your answer, it seems like a part of my question was cut off. What is important for the simulation is that the more the wheel deforms the more the contact area with the ground increases. Is there however a way to simulate that when using a distributed load? e.g. a distributed load that has a non-constant area.

    Due to larger deformations the internal structure will also mean the wheel will have self contact during deformation, is that automatically taken into account for the simulation?

  • GTT Adam
    GTT Adam
    Altair Employee

    Hi @Tim2746083295,

    In that case you can still include the contact plate in order to include the changing contact area, as I may have over-simplified the original setup.

    If you anticipate the structure coming into contact with itself, then you may want to partition the body so that the relevant faces can be assigned to contacts.

    Hope this helps!

    Adam Reid

  • Tim2746083295
    Tim2746083295 Altair Community Member

    Hello Adam,

    Once again thanks for your reply. I unfortunately still don’t get the results I want, maybe I am missing something. It seems like the problem should be relatively simple, so ill break down what I did below.

    Structure Analysis

    The first approach is using the structure analysis.

    image-caa1310383c59-d3e7.png

    I have a cylinder, free in vertical direction on a fixed ground. Force is applied at the center of the cylinder. Now I want to observe the deformation when contact between ground and cylinder are taken into account. Running the simulation will show the following:

    image-60b43e1d25af7-2bd4.png

    Here the cylinder will penetrate the ground. I tried with many different setting in the ‘contacts’ tool (e.g. changing tolerance & clearance) but no simulation showed the cylinder colliding with the ground.

    Motion analysis

    The same situation but for motion analysis, where the contact is a motion contact between cylinder and ground will run, but does not allow me to look at the deformed cylinder results. All I observe is the cylinder penetrating the ground with a couple mm at the start of the simulation. Here I’ve also tried playing around with the motion contact settings, but no results.

    image-13bb38b98159e-7d9d.png

    Analyzing the part after the motion does show a deformation plot, but not the deformed object I want to see.

    Which of the two approaches should I follow, and what do I change to get the desired result?
    Kind regards,

    Tim

  • GTT Adam
    GTT Adam
    Altair Employee
    Answer ✓

    Hi @Tim2746083295,

    For the Motion analysis, you will need to convert the cylinder to a Flexbody in order to allow the body to be deformable. The compression of the body can then be viewed using the Flexible Body tool:

    image.png

    If you are experiencing penetration of contacts, then they likely are not stiff enough. You will need to review the contact parameters and increase the stiffness values.

    It looks like the contact size in the Structural Analysis does not dynamically update, hence the deformed shape.

    I quickly mocked up an example using the Motion Analysis. Please see the results below:

    inspire_wheel_compression.gif

    This should be suitable for a preliminary analysis, such as to find out if there will be spoke contacts or general compression amounts. The simulation can be improved using smaller mesh sizes, stiffer contacts, etc.

    If you are trying to use this model for things like durability or lifecycle analysis, I would still avoid using this Flexbody method as flexible bodies are not suitable for large-scale deformation, hyper-elastic materials like rubber, and local plastic deformations, as the superposition of modes only applies to linear systems.

    Hope this helps!

    Adam Reid

  • Tim2746083295
    Tim2746083295 Altair Community Member

    Hello Adam,

    Thank you very much! This is exactly what I needed. I managed to get it working. It turned out one of the problems I had was that the joints/supports were flexible instead of rigid.

  • GTT Adam
    GTT Adam
    Altair Employee

    Hi @Tim2746083295,

    No problem! Glad I could help.

    Thanks,

    Adam Reid

  • Tim2746083295
    Tim2746083295 Altair Community Member

    Hello @GTT Adam,

    I made some steps in the simulation process but am currently stuck on a new problem. For the self contact of the flexible wheel I want to use the partition tool, each partition can then be assigned a separate motion contact with the part it will come in contact with during deformation.

    When partitioning the part that I partition should still be fully connected to the main body of the wheel. In my simulations, again a very oversimplified model to understand the process, I have connected the parts using locked joints. This does seems to be working as the parts are still connected, but the deformation is not what it should be. The joints put some kind of extra constraint on the partition which prevents the two parts as acting as one body where the deformation plot would be continuous.

    image.png

    How can I change the connection between main body and partition to have them act as one, but allow for defining self contact within the model?

    Kind regards,

    Tim