[LB7 (OP)]

The blue color is small balls. Each ball is attracted by a spring to the green line. Each spring attract from the green line to a ball. Like that I have the buoyancy like gravity can do in a container full of water. The torus can turn around the blue center. The device turns around the red center at the same angular velocity: the device is always like I drawn.

The grey part is attracted by springs from the green line of the other container. I can't attract all the part from one container to the other torus but enough for change the direction of the torque. With the springs I attract the same volume of the grey part that is inside the container to cancel all forces on the green line.

F1/F1' gives counterclockwise torque. F2/F2' gives a counterclockwise torque. The only problem: springs rotates the torus in the clockwise direction, so I will attract the torus 1 with the container 2 and attract the torus2 with the container1, like that the torque on the torus is lower and I think I can change its direction when I see the volume.

I can increase the hatching area.

[wolfekeeper]

no

Your device is very, very well described by Newtonian Mechanics, and there's a general proof that overunity is completely impossible in Newtonian Mechanics.

[wolfekeeper]

not even then

[LB7 (OP)]

I'm hatching the area where I exchange:

Container 1 attract the torus 2
Container 2 attract the torus 1

The springs attract all the blue area.

Before the green line attract the green torus and the red line attract the red torus. But now, the green line attract the red torus where the hatched area is (don't forget there are 2 torus) and the red line attract the green torus where the area is hatched. The green line attract the rest of the green torus (not hatched area), and the red line attract the rest of the red torus like that there is no force on the green line and on the red line. Like that I don't have any force on the red line and on the green line.

I drawn the container1 and the torus1 alone and the container2 and the torus2 alone to look the differences.

The device turns around the red center, it is always like I drawn: it's important to keep constant the length of the springs. The springs don't lost any energy.

[LB7 (OP)]

The first image shows the device in the side view.

The second image shows how is the attraction from Container1 to Torus2 and from Container2 to Torus1.

The third image shows how the device turns around the red center.

The springs keep their length constant, the device turns around the red axis but all turns at the same angular velocity.

I calculated the torque:

Code: [Select]


from math import *

N=2000
c1=0 #number of points of the torus1 inside the container1 or the surface if divised by N²
c2=0 #number of points of the torus2 inside the container2 or the surface if divised by N²
c3=0 #number of points of the hatching area or the surface if divised by N²
couple1=0.0 #clockwise torque on the torus1 (from springs) IF all part is attracted by the green line
couple2=0.0 #clockwise torque on the torus2 (from springs) IF all part is attracted by the red line
couple3=0.0 #counterclockwise torque on the torus1 or the torus2 due to the presence of the hatching area

Ra=7.0 #inner radius of the tore
Rb=8.72 #outer radius of the tore
dx=Ra+Rb #the center of the torus1 is (0,0), the center of the torus2 is (dx,dy)
dy=4.92

for i in range (0,N) :
for j in range (0,N):
x=i*(Ra+Rb)/N
y=j*dy/N
if x**2+y**2>Ra**2 and x**2+y**2<Rb**2 and x>Ra and x<Rb and y>0 and y<dy:
c1+=1
couple1+=y*x
if (x-dx)**2+(y-dy)**2>Ra**2 and (x-dx)**2+(y-dy)**2<Rb**2 and (x-dx)<-Ra and (x-dx)>-Rb and (y-dy)<0 and (y-dy)>-dy:
c2+=1
couple2+=abs(y-dy)*abs(x-dx)
if x**2+y**2>Ra**2 and x**2+y**2<Rb**2 and x>Ra and x<Rb and y>0 and y<dy and (x-dx)**2+(y-dy)**2>Ra**2 and (x-dx)**2+(y-dy)**2<Rb**2 and (x-dx)<-Ra and (x-dx)>-Rb and (y-dy)<0 and (y-dy)>-dy:
c3+=1
couple3+=y*x

print c1
print c2
print c3
print couple1/N**2
print couple2/N**2
print couple3/N**2

The results:

Couple1=1.169
Couple2=1.169
Couple3=0.895

The Couple1 is the torque if I have attracted the torus 1 with the green line, like the hatching area gives 0.895 this would say the rest of the area attracted by the green line is 0.274 clockwise so the torque on each torus is counterclockwise of 0.895-0.274=0.621


Energy is constant: qt1.png


Energy not constant: sr2.png

[LB7 (OP)]

i2.png (16.38 kB . 1052x552 - viewed 3224 times)With one torus and one container it must be ok but the torus must be in 2 parts:

- the inner wall and the outer wall receive the pressure from balls, the sum of forces are transmit to the center of the torus, it gives F1, F1 gives a counterclockwise torque
- the inside torus is free to turn in the walls, the inside torus will receives a clockwise torque but lower

The container receives a counterclockwise torque from F2.

I calculated the torque with:

Inner radius=7
Outer radius=8.72
height ot the container=4.92
Pressure=height

The torque from F1 is 17Nm
The torque to the green torus is -11.8 Nm
The torque to the container is 6 Nm

The sum is not 0.

At start, all parts of the device turns counterclockwise at w around the red center. I apply my method, the green torus slow down but the device accelerates more.

If I add a motor on the blue axis, I can compensate the torque from the springs, like that the springs keep their length. The motor need to give an energy but less than the torque from F1