I tried to model a simple flywheel & gimbal and observe the gyroscopic forces, but I can't seem to get the magnitude of gyroscopic torque that I expect. I expect the magnitude of the gyroscopic torque to be Izz * spin rate * gimbal rate. However, when I increase the spin speed by 2x, I've been getting torques out that are closer to 10x greater. My torque plots also look something like this. Is there a setting I need to change to enable calculation of gyroscopic torques?
We have also recently investigated problems with the accuracy of gyroscopic torques. Basically, the solver adds additional gyro-stabilization torque in certain cases in order to make the solution more stable. But in cases where things are spinning quickly, these torques can be far larger than what should exist for those models. We're currently working on a way to disable these forces when they're not needed or when accurate torque measurements are needed, but this is not yet possible.
So the short answer is that gyroscopic forces on fast spinning objects are currently not accurate in Vortex, but we should have a way to fix it for most use cases in the future.
In the upcoming (early 2020) Vortex Studio, parts have a new input field
named "Disable Gyroscopic Stabilization". It is unchecked by default.
Gyroscopic stabilization is required for the stability of unconstrained angular degrees of freedom; the system can become become unstable without it. In certain cases where a constrained body is spinning quickly (such as an engine drive shaft) these forces can become unrealistically large and the gyroscopic stabilization should be disabled.
Which we believe applies to your current use-case. You could therefore check this box for the relevant part.