Consider adopting the IMPACT contact model from MSC Adams
https://essay.utwente.nl/62109/1/BSc_J_Giesbers.pdf It differs from the current implementation in two ways:
- An exponent parameter for creating nonlinear springs
- A damping parameter that is gradually applied over a maximum penetration depth
Especially the way a damping force without any discontinuities is created could be useful. It might improve the stability of the dynamic simulation.
Edited by Stefan Pfeifer