34m and 5MW spinning, structures only has large reaction torque

[kevmoor]

Same mesh and sectional properties as used for the 34m Campbell diagram (which matches very well) fixed bottom, fixed top but allowed to rotate about z. It is returning a large reaction torque at the base (~170 kN), which is a little higher than the mean aerodynamic force (if it was on for the 34m at 34RPM). I’ve recreated the problem using a different solver/code, geometrically exact beam (GXBeam), using the same setup for mesh, properties, conditions, and am getting a similar ~150kN reaction torque. From a lay man’s perspective it seems like we should be able to spin a top (or turbine) in a vacuum and there be basically no reaction torque at the base (in a rotating frame of reference) but perhaps there is something I’m missing since both codes are so similar? I’ve increased and decreased RPM to ensure we’re not on a natural frequency. Both codes do match the experimental 34m Campbell diagram quite well so I have confidence in the mesh and sectional properties.

This was temporally resolved by zeroing out the sectional properties cg offset. It may be that the CG offset should be reversed for the other blade and/or rotated by the blade global twist (due to its azimuthal location) such that the effects zero out at the base. The force effects from these small offsets should be small, keeping them zeroed out for now shouldn't introduce more than a few percent error, keep issue open until we try rotating the cg coordinates.

It may also be that the CG offset equation set sign should be reversed.