Sum of torque

[lumen]

I think the design could be made less complicated by reducing the working principal to just two working components.
1: centrifugal force
2: Roberval action
I have done some testing on the principal and have not found any cause for the device to fail so I am now building a test model.
There must be a condition to cause the device to fail but it's simply not apparent.

[EOW]

What is Roberval action ? The balance ? there is no gravity here.

Yes, sure for a non continous machine it's possible to test with just 2 circles. Better efficiency with several circles. But do you tested with liquid or gas ? It can be tested with a "velocity" higher than the speed of sound. The "velocity" is the velocity of the red line. Where are you discussed about that ? The speed of sound in water is 1500m/s and for test, it's not easy. I drawn lines of equal pressure from another circle when the door is open. The force F from the red line don't give the same torque, it's logical. But, this additionnal torque don't work the same in one turn (of 2 circles) and it works a small time, "waves" (curved lines black, red, magenta and blue) move at 1500 m/s

[lumen]

Roberval action is simply the mechanics that maintains the orientation as the chambers rotate around the circle.
If they were aligned to north before starting, they would remain pointing north while rotating.
In your drawing 1-1 is pointing the same direction as 2-2 and 3-3 then 4-4 , 5-5, 6-6. They point the same direction as the disk rotates.

This is what forces the gas (or liquid) to move from one chamber to another by centrifugal force.
There is always an apparent gain in energy even at slow speeds.

[EOW]

Have you already talked about this idea somewhere ? Have a link for Roberval ?

Sure, it's work at any velocity but:

1/ for test with sensors, you need to detect something that run at 1500m/s
2/ with 2 circles, the device like I drawn don't work, in one turn the sum of work is 0, the door must be closed some steps and open others steps

With a high velocity, the waves  don't have time to stabilize them, so it's better. For me it's the delay that can give energy and it's logical because the angular velocity of circles can be very high. If the device lost energy when the liquid moves (water is compressible so a few liquid will move), this energy is near constant, but the angular velocity can be 10 times or 100 times higher.

The centrifugal force is mv²/R, so with a big v and a big R, the centrifugal force is not high, but the distance that circle run is high, so it must be ok. Tell me more about your tests, on a software ?

[lumen]

Your plan to use a Gas and extract energy is not exactly what I was doing but there are many ways to extract energy from the two basic systems combined.

I plan to generate electrical current directly from centrifugal force using a roberval system on a rotating disk.
The tests I have been doing are to find out if the torque from the centrifugal force can in some unknown way apply back against the rotation of the disk generating the centrifugal force.

Torque on a roberval device cannot apply in any way to cause rotation of the disk but centrifugal force is radial and some things change because of the vector of the force has vector differences over the working area involved.

Regardless of these facts there is a way to correct for that problem so again I'm left searching for the cause of failure in such a device.

It will be interesting to know the exact cause of failure for such a device because it appears so elusive.