Scooter investigation | Tire contact (bump) modeling and rayleigh damping questions

[Qiyuan Zhou]

Hi

I've spent the past few weeks working on creating a model using Project Chrono to place bounds on a rather catastrophic failure mode in an electric scooter I have purchased. The project has given me some reasonable results, and is a more in-depth description is available on the project's Github page: https://github.com/keonjoe/Nami-Simulator

There is still some work to be done on the correlation side and tweaks to the contact model to be made. In this regard, I do have a few questions:

1. Could somebody please comment on how accurate the standard Hertzian contact model has been for handling the vertical forces (i.e. from hitting a sharp bump) from tire-road contacts? Are there any recommendations for how to tweak the contact model to better approximate the behavior of a tire over a bump if the Hertzian contact model is not a good fit for this application?
2. I noticed that when using a lower stiffness value for the ChElementBeamIGA elements I used to create a part in my model that some spurrious low frequency oscillations are present. When I was working at GT helping customers troubleshoot these types of issues in their models, the solution was usually to increase the alpha term in the material Rayleigh damping model. However, I noticed in the comments in the code that it is not possible to apply the alpha Rayleigh damping term. Is there an explanation for why this cannot be done for ChElementBeamIGA? Surely there is a mass matrix being assembled at some point in the solution process on which the alpha term can be applied.

Thank you in advance for your help!

Qiyuan

[Luning Fang]

Hello Qiyuan,

For the question regarding the Hertzian contact, if you have a good estimate of the material properties, such as stiffness and damping, it should model the physics decently. Though hard to measure the accurate​ contact force, what we do know is, the Hertzian contact model leads to macroscopic behavior of the dynamics system that can be measured and have been validated. Note that you need a small step size for the solver to be stable, in your case, it's a box-triangle mesh contact, so it could be time consuming.

The good thing about Chrono is, you can have a user-defined contact model for your application. Take a look at the class ContactForce in demo_MBS_callbackSMC.cpp, you can implement whatever you want for your tire-bump contact, it does not have to be Hertzian or Hookean.

Thank you,

Luning