Solid gyroscope and sum of energy

neufneufneuf

I take 2 half cylinder mass m, diameter 10 cm, they are away from each other of 10 m. I move them at velocity v for formed a gyroscope like image shows (one half cylinder move at v, other at -v). This need energy 2*1/2mv². Now it's a gyroscope, it turn at w. Centripetal forces F create a torque, this give a precession at small angular velocity Wp. Wp is perpendicular to w, so velocity of each particle is bigger (sum of vectors w and Wp). Now I break the gyroscope (in theory, in a lot of particles), I recover energy and it's bigger than 2*1/2mv² ? So what's wrong in my thinking ?

neufneufneuf

Even the energy is small. It's possible to put this system on another gyroscope. With a Powerball, it's possible to accelerate the disk with a torque, so if this system change the torque of another powerball, the disk must increase its energy like time. It's an ocillator over an oscillator, the response is t*cos(t).

Or better, put an asymmetric ball inside a Powerball, the torque from centripetal forces must increase angular velocity of ball. The torque is block to wall a part of angle and works a part of angle.