Learning Forum

[Carlos Valle Noguera]

Hi everyone,

I'm doing my academic thesis in the simulation of a dental impact but I can't figure out why the bodies keep behaving like rigid bodies and don't make contact with one another, just pass through.

 

I've tried with changing the pinball region, increasing the contact region, changing the steps on the analysis settings to no time integration until contact has occured, introducing a contact stabilization damping, the adjust to touch option and other things i've found that may have solved the problem but it didn't. 

I dont' have any problems sharing the project and the geometry file in order to solve to problem.

I'd greatly appreciate some help, thanks in advance,

 

Carlos Valle

 

[dlooman]

A possible explanation is that the surfaces are over-penetrating before contact is established.  Depending on the time step size, it's possible for the surfaces to be not yet in contact at one time point and then have passed through each other at the next time point.  To obtain a good solution at the time of impact it's usually best to model the geometry in contact and then give the appropriate initial velocity to obtain the dynamic effect of impact in conjunction with using very small time steps.

[Carlos Valle Noguera]

Okay I will try that and update on the results, and also would you recommend using transient structural or explicit dynamics? as the impact is a low velocity one I'm not sure wich one to use

[dlooman]

I'm biased, but I prefer transient structural for impact.  Explicit is good too of course.

[Carlos Valle Noguera]

Hello Dave,

I've tried the simulation with transient structural and explicit dynamics, but I've only managed to get results in explicit dynamics. With transient structural the implant deforms heavily and strangely when in contact and the final result doesn't make any sense. I've already defined the contact as frictionless and adjust to touch and with a very min time step of 5,e-008. Would like to know if you would consider taking a look at the model to chech for any erros I may have done.

I share some photos of my model. Thanks in advance.

Carlos Valle

 

[dlooman]

Yes, you could email me at dave.looman@ansys.com

[Carlos Valle Noguera]

The problem I had was not changing to "true scale" on the results section on my last attempt and with the settings used on the pictures. 

Thanks for all the help and feedback Dave