Trying to answer the questions a little better. . .
I would have thought that after setting the center point for the rotation, which needs to happen anyhow. . . from that position x red direction would act similar to what happens in the RotX of a dynamic component.
In messing around with the Rot values of a dynamic component I start to see a different reality which you're trying to explain to me.
While I can get the first angle of rotation to happen as I would imagine it, the 2nd rotation is way off from expectations.
. . . But still obtainable I would think, there just has to be a little memory build into the rotation sequence. so that whatever the settings were prior to the first action, those still temporarily hold over for the 2nd rotation as well. allowing 2 directions at a time.
and yes folks what I wrote, actually makes sense to me,. . . I don't doubt that this is crazy talk for someone who really knows this stuff.
I guess it's a good thing that I'm nowhere near a job site that deals with the geo synchronous orbits of satellites. which with my input would be back down on the ground in no time at all. Hopefully not on top of someones house.