Attached above is an image of this behavior. The red part is a heavy part without massless on, the blue part has the same mass as the red part but massless is on, and the yellow part is a near massless part demonstrating how a massless part should behave.
I have to keep the mass of my steering hinges high on my vehicles so that they work and do not unalign or stop working completely like the attachments on the yellow part pictured above, however this is impacting the torque of the vehicles greatly because of massless part behavior.
Attached is the massless test model: MasslessTest.rbxm (5.1 KB)
Thank You
-Widgex
The issue here is just a simple misunderstanding. The massless property does not apply if the part is the root part of an assembly. This limitation ensures that the overall assembly has a non-zero mass and moment inertia so that the solver can successfully operate on it. More details can be found here:
and the limitation is noted in the documentation here:
The proper behavior can be demonstrated in the attached test by welding an additional box with mass to each of the existing parts. In the picture below, I tweaked the original test by orienting the hinge axis horizontally so that the motor force must resist gravity and I altered the force so that it is barely capable of keeping the main gray block horizontal.
Adding the massless block (blue) or the light block (yellow) has no effect on the equilibrium position of the root block while adding the heavy block (red) is enough to cause the root block to fall to the vertical position. Hopefully that clears things up. Let us know if something still seems off as obviously vehicle dynamics is an important use case that we want to make sure we get right.
