TSDA handling in DoStaticNonLinear()

[Paf]

When running DoStaticNonLinear() on a system with TSDA, I assume the Newton-Raphson iterations would find an equilibrium TSDA length to match the forces/torques in the system, but it doesn't seem to change. Only kinematic constraints are resolved.

In attached pics,  changing the spring K changes the magnitude of forces in the system.

Using a SPARSE_QR solver to avoid any jitter.

Am I doing it wrong ? What's the proper method to accomplish this?

Thanks,

Pavel

[Paf]

Setting TSDA IsStiff(true) resolved the issue.

[Radu Serban]

I see you have already figured out the solution to this problem.

Indeed, the Jacobians necessary for nonlinear static analysis for a system including a TSDA are only generated if the force element is declared as stiff (in which case these Jacobians are generated to be used within an implicit integrator). This is obviously confusing, so I’ll look into making this less obscure for situations such as yours.

--Radu

Pavel

--
You received this message because you are subscribed to the Google Groups "ProjectChrono" group.
To unsubscribe from this group and stop receiving emails from it, send an email to projectchron...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/projectchrono/f31e81b1-a583-4528-99e3-f99d893c215bn%40googlegroups.com.