Please excuse my English...
Let's say we have a 10m x 10m x 10m water tank filled with water.
After a little search i found that the pressure on the sidings at 9-10m depth would be approximately -> 1 bar => 100 KN/m2.
Now let's assume that we sink 400 spheres with a diameter of 1m in the tank (100 every 1m of depth (with a surface of 10m x 10m = 10 m2))
and we've placed a funnel (with some holes on its sidings to "breathe") at the top, which guides the spheres into a (non airtight) tube 1m diameter.
The tube will then guide the spheres, outside and at the bottom, of the tank (at 9-10m depth).
If we assume that the spheres are made of steel 2mm thickness, their weight would be approximately 50Kgr, and their volume 500lt. So their buoyancy would be 4.5KN.
If the tube was already filled with spheres, i end up that the force at the exit point would be 4.5KN x 400spheres - friction => 1,800KN - friction.
I believe that the friction could surely be less than 1,700KN. Right???!!!
So if I'm right, the "force in" would be many times more the "force out" (pressure).
And if I'm right this could work everywhere even in a spaceship (through centrifugal)...


