Simscape Rotational Friction – Instability Issue in Joint Simulation

- Breakaway friction torque: 20 Nm
- Coulomb friction torque: 10 Nm
- Breakaway friction velocity: 0.1 rad/s
- Viscous friction coefficient: 50 Nm·s/rad

- Why this divergence is happening
- Whether there are any best practices or recommended solver settings when using the Rotational Friction block in this context
2 commentaires
Hi @ donghoo,
I do agree with @Yifeng Tang’s comments, however, the only advice I can give is that the instability is probably related to the discontinuous nature of Coulomb friction. Even without Matlab Simulink access, here are some general approaches that often help:
1. Solver settings: Try ode15s or ode23s (stiff solvers) instead of fixed-step
2. Regularization: Most friction blocks have options to smooth the velocity discontinuity
3. Initial conditions: Make sure your initial joint velocities aren't exactly zero
4. Step size: 0.001s might be too large for the friction dynamics - try reducing it
Please bear in mind that the key is avoiding the mathematical discontinuity at zero velocity that causes numerical issues. Hope this helps.
Réponses (0)
Voir également
Catégories
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!