Considering the typical situation of a rotating bicycle wheel held by one end of its axle by a rope tied to the ceiling: gravity torque is the time derivative of the angular momentum, and in this case is perpendicular to the angular momentum vector which makes it revolve without changing its magnitude (it precesses).
My question is the following: the wheel stops turning eventually because of the friction in the string (it also slows down its own rotation because of friction in the bearing but let's ignore it), but could you clarify how, step by step? Is it because the gyroscope creates a torque on the string axis, but it's reduced by the friction torque, and as a result the friction torque translates back into a residual "gravity torque" (in other terms, the gyroscope "gives up" on a certain amount of gravity torque it can't counteract because of the string)?
To progress toward my actual problem in several dimensions, what if the friction torque is in fact torque given by another gyroscope on the same solid?