Hinge constraint not converging

protontim

Hi Victor,

I made a model of the my rotor with 3D non-linear elements and added the centrifugal forces with a script I wrote. That simulation works very nicely (thanks!). There was high stress at the mounting point of the rotor, where the faces were fixed. I wanted to see how a rotor on a "hinge" would work so I picked two horizontal points at the base of the rotor and fixed their displacements in all three directions, but did not fix their rotations. I ran the model, but it would not converge. Any ideas how to get this thing to converge? As I mentioned, it converges just fine when I fix the face that I now want to have on a "hinge".

Thanks,
Tim

Victor

A hinge is a little tricky to converge because it's underconstrained at the first iteration. A couple of ways are:

1) Use time steps (specify time step info in the global properties) and apply the force to stiffen the structure at a time before any force that tends to rotate the hinge. There's an example of this technique in the SaggingCable.liml sample file. I don't think it will work here though unless the centrifugal force is perfectly aligned so it applies no moment about the hinge.

2) Put a soft spring somewhere to gently constrain the rigid body rotation. In case the spring is too weak and it still doesn't converge, you might need to ramp up the force gradually over several time steps.

In general, if it doesn't converge and you can't work out why, reducing the forces or ramping them up from zero can often show the problem. It'll usually converge for a few time steps then fail at some point and you can look at the final solution which might reveal some unexpected deformation.