Gravitywheel with buoyant compensation

[Low-Q]

Hi there,


Just got an idea. I hope you guys see what I'm not seeing - conservation of energy.
In the images I have two scenarios of the same submerged wheel.


I use buoyancy and weights to compress and expand air compartments in the ends of a tube. When the tube is horizontally, the math is easy. The right hand side wants to move up due to buoyancy.
The left hand side has less buoyancy, and wants to move down.
As you can see, there is two pistons on each side. the upper piston to the right, and lower piston to the left is forced up due to 1kg buoyancy of the attached air pockets.
On the other hand, there is a weight, including water displacement, weights 1kg, which force down the attached pistons on the left and right side.


However, when this tube is turned vertically, the higher water pressure at the bottom will force the bottom pistons towards eachother. In order to compensate for this force, I have attached a compensation.
This compensation is attached to each rods, red and blue, so when the tube is vertically, the compensators are forcing the pistons to stay in the same position.


So, the idea, is that there is a buoyant force between the pistons on the right hand side will generate torque CCW...


Well, it is late, and I'm tired. So I cannot figure out where the conservation of energy is. It must be some place. So if you got time, please help me find it. If you can't, I will start building a prototype - a small one.


EDIT: It should be 10Nm torque. Not 20Nm, as the radius in 0.5m.


Good night.


Vidar

[Low-Q]

Finally figured out where the conservation of energy goes.
When the tube is horizontal, and the pistons weigths turns the compensator rods slightly a few degrees, the weight and buoyancy on the rods moves slightly to each sides, and off axis of the rotor.
The weights on that compensator (Upper and lower mass in the first image) must be enough to counterfight the pressure difference over a given area of the air compartments in the tube when the tube is vertical.
The surface area on the pistons require quite much force to be kept in place, so the compensator weights is also heavy and very buoyant.
This very mass displacement that is placed slightly off axis, generates a torque that is exactly the same as the torque given by the right hand air compartment in the tube.


See image below for reference.


So, there you go. Sorry for wasting your time.


Vidar.